Emanuel Tov
The Greek & Hebrew Bible
Collected Essays on
the Septuagint
THE GREEK AND HEBREW BIBLE
SUPPLEMENTS
TO
VETUSs TESTAMENTUM
EDITED BY
THE BOARD OF THE QUARTERLY
H.M. BARSTAD — PHYLLIS A. BIRD — R.P. GORDON
A. HURVITZ — A. van per KOOJ — A. LEMAIRE
R. SMEND — J. TREBOLLE BARRERA
J.C. VANDERKAM — H.G.M. WILLIAMSON
VOLUME LXXIil
THE GREEK
AND
HEBREW BIBLE
Collected Essays on the Septuagint
BY
EMANUEL TOV
BRILL
LEIDEN : BOSTON : KOLN
1999
This book is printed on acid-free paper.
Die Deutsche Bibliothek - CIP-Einheitsaufnahme
[Vetus testamentum / Supplements|
Supplements to Vetus testamentum. - Leiden ; Boston ; Kdln :
Brill
Friiher Schriftenreihe
Rethe Supplements zu: Vetus Testamentum
ISSN 0083-5889
Vol. 72. Tév, ‘Immanwtél: The Greek and Hebrew Bible. - 1999
Tév, Immaniél:
The Greek and Hebrew Bible : collected essays on the Septuagint /
by Tov. - Leiden ; Boston ; Kéln : Bnill, 1999
(Supplements to Vetus testamentum ; Vol. 72)
ISBN 90-04-1 1309-6
Library of Congress Cataloging-in-Publication Data is also available
ISSN 0083-5889
ISBN 90 04 11309 6
© Copyright 1999 by Koninklyke Brill NV, Leiden, The Netherlands
All nghts reserved. No part of this publication may be reproduced, translated, stored in
a retrieval spstem, or transmilted in any form or by any means, electronic,
mechamcal, photocopying, recording or othenerse, unthout prior written
permission from the publisher.
Authorization to photocopy items for internal or personal
use ws granted by Brill provided that
the approprate fees are pard directly to The Copyright
Clearance Center, 222 Rosewood Drue, Suite 910
Danvers MA 01923, USA.
Fees are subject to change.
PRINTED IN THE NETHERLANDS
DEDICATED TO THE MEMORY
OF TWO VERY CLOSE RELATIVES
WHOSE LOSS IS GREATLY FELT
JUDA KOEKOEK 9”1 (1914-1998)
DEBORAH M. E. KOEKOEK °”} (1971-1998)
CONTENTS
PACE & cca ale kosrite acai. toaatar ave aor aca leche an aie eae eee a ca eet XI
SOUTCCS soak as Gs EN aioe Holds bie Ssh el Raa Gad aa ee Gees he Xi
Editions of Textual Sources:..2.:64 thsaeerdny le hee yew tab hetens xvi
Periodicals, Reference Works, and Serials ...................00. XLx
Bibliographical Abbreviations ............. 0.0.0.0. ee eee eee eee Xxii
I. GONCT A SEMA ICS: 424 eget tea Baath Goa Ped ete beech dae ny Se
IT.
MY.
™N
12.
13.
The Rabbinic Tradition concerning the ‘Alterations’
Inserted into the Greek Translation of the Torah and
Their Relation to the Original Text of the Septuagint ...... 1
The Fifth Fascicle of Margolis’ The Book of Joshua in Greek 21
A Computerized Database for Septuagint Research........ 31
Glosses, Interpolations, and Other Types of Scribal
Additions in the Text of the Hebrew Bible................ 53
Eine Jorajur dem Konie Tanna ntc08 tvit b ede cece Senos: iD
LOMICORT ID « oops GE Reh dk oi at Benet eh ent BE tea a Saeed 83
Three Dimensions of Words in the Septuagint............. 85
Some Thoughts on a Lexicon of the Septuagint ............. 95
‘Greek Words and Hebrew Meanings’.................... 109
TPQHSIATION: TeChMIQUe QUE: -EXCSESIG sd iis ij sierad dow dagwehe 129
Compound Words in the Septuagint Representing Two or
More: Hebrew W Ords sis one honmen de tae eoatewere es 131
. Midrash-Type Exegesis in the Septuagint of Joshua....... 153
. Loan-words, Homophony, and Transliterations in the
SEP MAINE 6. wa adits erate eer es Sree eee A ae tee oe 165
The Impact of the Septuagint Translation of the Torah
on the Translation of the Other Books.................... 183
The Representation of the Causative Aspects of the
PLP) ANS SEP PUA I cote: tsa bes die Sesh Space Rees eh 195
VITI
14.
15.
16.
dle
18.
IV.
19.
20;
ye
22
23.
24.
hey
26.
oy
28.
Piel
CONTENTS
Did the Septuagint Translators Always Understand
Theirebrew Vext? t2c52-po4ctadeanaweeee ee whee eene ees
Computer-Assisted Study of the Criteria for Assessing
the Literalness of Translation Units in the Septuagint .....
The Nature and Study of the Translation Technique of
ida esate, =) 8) 10 124 ba Ge apan aa ee ere ere gre eae tee em eee
Renderings of Combinations of the Infinitive Absolute
and Finite Verbs in the Septuagint—Their Nature and
DIS DUOM 5 daa Picseds bt cease eee ei eee els
Theologically Motivated Exegesis Embedded in the
DePIUaCING vette eh ee heey ee th eokithatenaKeha
The Septuagint and the Textual Criticism of the Hebrew
BS OE er seae hc tend creates heap a se Babhena dew: fete pian Aa petal a ean
The Textual Affiliations of 4QSam?#................00505
The Contribution of the Qumran Scrolls to the
Understanding Of the Sepa gints.c.s ice taka steer a siemens
Interchanges of Consonants between the Masoretic Text
and the: Viorlige ot ne Sepilapint 32 nie anew ee eee ite ws
The Septuagint and the Literary Criticism of the
TA CRT EU TBIDIC Gist atk oc, We Arie a tate bh es gee ne ete
Exegetical Notes on the Hebrew Vorlage of the
Beptiiagmtof feremualy2/(94) tnd avadanein wee ae ages
The Composition of 1 Samuel 16-18 in Light of the
Dep UG GIAY etki hence eke eee edn ee ea uote aes eaee
The Literary History of the Book of Jeremiah in Light
OF NS Pex ial IS IOny vases atoy edna A Sit Aes acs oto
The Growth of the Book of Joshua in Light of the
Evidence Of he Sepitas int o24axS ab tanenduretin tage kata Sa
Recensional Differences between the Masoretic Text and
Ine Sepilagint( Ol EZeKICl bieiwt je .bibe.6 wade ee kb S Raed os
Some Sequence Differences between the Masoretic Text
and the Septuagint and Their Ramifications for Literary
CHINCISIl 3204, 0a0eeee ¢obc el eucitieee svedabekain kaa
Recensional Differences between the Masoretic Text and
ihe Sepa eintol Proverbs, irae i otacas ecaae eee ous
Different Editions of the Song of Hannah and of Its
INarratiVe*P TANG WOLK <n cciaatiaesdaacoeh eee cients
203
VI.
30.
one
32
33.
34.
Jo:
36.
evi
38:
CONTENTS
RECISIONS OF THE Sep AI a erst 5 ta Bid and 0a, Ba deee ORE e oes
Pap. Giessen 13, 19, 22, 26: A Revision of the Septuagint? . .
Lucian and Proto-Lucian—Toward a New Solution of the
FOP ET 24Ginfan oie tases att sao seine ee aeareade
The Methodology of Textual Criticism in Jewish Greek
Scriptures, with Special Attention to the Problems in
Samuel—Kings—The State of the Question...............
Transhterations of Hebrew Words in the Greek Versions—
A Further Characteristic of the kaige-Th Revision?......
Une inscription grecque d’origine samaritaine trouvée a
MINGSSAlONIG Ue Grctinicdvte se Ee Laine dd ane oe Gea aed ale
The Relation between the Greek Versions of Baruch
The Textual History of the Song of Deborah in the A Text
GENE Se DUA CUI chore. ica ca ena id nein Sie, Sarak oma aucune deans
The ‘Lucianic’ Text of the Canonical and Apocryphal
Sections of Esther: A Rewritten Biblical Book............
The Septuagint Additions (‘Miscellanies’) in 1 Kings 2
(BIREICNS 2) jas anaitttdesug wosten se Gene ain dan toe ede
IX
501
PREFACE
My interest in the Greek translation of the Bible was first kindled by a
course given by the late Professor I.L. Seeligmann in the second year of
my B.A. studies at the Hebrew University (1962), and ever since this
area has remained one of the main focuses of my scholarly interest. The
six section heads of the thirty-eight chapters reflect the various
aspects of my investigations on the Septuagint: lexicography,
translation technique and exegesis, the Septuagint and the textual
criticism of the Hebrew Bible, the Septuagint and the literary
criticism of the Hebrew Bible, and the revisions of the Septuagint. The
first section contains several general studies.
This volume contains the great majority of my articles devoted to
the Septuagint, with the exclusion of introductory articles on the
Septuagint in encyclopaedias and handbooks, articles which were
Vorarbeiten for my book The Text-Critical Use of the Septuagint in
Biblical Research (Jerusalem Biblical Studies 2, 8; First edition 1981;
Second edition, revised and enlarged, 1997; Jerusalem: Simor), and most
articles related to the CATSS project (Computer Assisted Tools for
Septuagint Studies).
The thirty-eight studies collected in this book were originally
published between the years 1971 and 1997 in various journals and
monographs. For a detailed list of these sources, see pp. Xili-x1v
All the studies included in this monograph have been revised,
expanded, or shortened, in some cases considerably. In two cases two
articles have been combined to form a new one. In all instances I tried to
integrate in the analysis studies which appeared subsequent to the
original articles. In some cases the discussion could be shortened by
referring to two monographs: my aforementioned book TCU and Textual
Criticism of the Hebrew Bible (Minneapolis and Assen/ Maastricht:
Fortress Press and Van Gorcum, 1992).
I owe a debt of sincere gratitude to several people who helped me to
create the present version of my collected writings on the Septuagint.
Steve Daley kindly read the complete manuscript critically, and
remarked on many details. Janice Karnis improved the English of half
of the chapters in the book. In the technical production of the volume I
XII PREFACE
was greatly helped by my children. While the more recent articles
were available in electronic form, most of the articles were not. These
articles were type-set on the computer by Ariel and Ophirah. The
manuscript of the complete volume was meticulously proofread by
Amitai and Ayelet. Most references were also rechecked in the sources.
I am grateful to Koninklijke Brill of Leiden, and especially to Mr. H.
van der Mei, for their encouragement in the publication of this
monograph and for the skilful production. The editors of the
Supplements to Vetus Testamentum are to be thanked for including this
monograph in the series.
Jerusalem, Pesach 5759 (April 1999)
Emanuel Tov
SOURCES
1. “The Rabbinic Tradition concerning the ‘Alterations’ Inserted into
the Greek Pentateuch and Their Relation to the Original Text of the
LXX,” JS] 15 (1984) 65-89.
2. “The Discovery of the Missing Part of Margolis’ Edition of
Joshua,” BIOSCS 14 (1981) 17-21; “The Fifth Fascicle of Margolis’ The
Book of Joshua in Greek,” JOR 74 (1984) 397-407.
3. “Computer Assisted Alignment of the Greek-Hebrew Equivalents
of the Masoretic Text and the Septuagint,” in: N. Fernandez Marcos
(ed.), La Septuaginta en la investigacion contemporanea (V Congreso de
la IOSCS) (Textos y Estudios “Cardenal Cisneros” 34; Madrid 1985)
221-242.
4. “Glosses, Interpolations, and Other Types of Scribal Additions in
the Text of the Hebrew Bible,” in: S.E. Balentine and J. Barton (eds.),
Language, Theology, and the Bible—Essays in Honour of James Barr
(Oxford 1994) 40-66.
5. Review of: G. Veltri, Eine Tora fiir den Konig Talmai— Unter-
suchungen zum Ubersetzungsverstdndnis in der jiidisch—hellenistischen
und rabbinischen Literatur (TSAJ 41; Tubingen 1994), in: Scripta
Classica Israelica 14 (1995) 178-183.
6. “Three Dimensions of LXX Words,” RB 83 (1976) 529-544.
7. “Some Thoughts on a Lexicon of the LXX,” BIOSCS 9 (1976) 1446.
8. “Greek Words and Hebrew Meanings,” in: T. Muraoka (ed.),
Melbourne Symposium on Septuagint Lexicography (SCS 28; Atlanta,
GA 1990) 83-125.
9. “Compound Words in the LXX Representing Two or More Hebrew
Words,” Bib 58 (1977) 189-212.
10. “Midrash-Type Exegesis in the LXX of Joshua,” RB 85 (1978) 50—
61.
11. “Loan-words, Homophony and Transliterations in the Septua-
gint,” Bib 60 (1979) 216-236.
12. “The Impact of the LXX Translation of the Pentateuch on the
Translation of the Other Books,” in: P. Casetti, O. Keel and A.
Schenker (eds.), Mélanges Dominique Barthélemy (OBO 38; Fribourg/
Gottingen 1981) 577-592.
XIV SOURCES
13. “The Representation of the Causative Aspects of the Hiph4l in
the LXX—A Study in Translation Technique,” Bib 63 (1982) 417-424.
14. “Did the Septuagint Translators Always Understand Their
Hebrew Text?” in: A. Pietersma and C.E. Cox (eds.), De Septuaginta,
Studies in Honour of John William Wevers on His Sixty-Fifth Birth-
day (Mississauga, Ont., 1984) 53-70.
15. “Computer-Assisted Study of the Criteria for Assessing the
Literalness of Translation Units in the LXX,” Textus 12 (1985) 149-187.
16. “The Nature and Study of the Translation Technique of the LXX
in the Past and Present,” in C.E. Cox (ed.), VI Congress of the
International Organization for Septuagint and Cognate Studies (SCS
23; 1987) 337-359.
17. “Renderings of Combinations of the Infinitive Absolute
Construction and Finite Verbs in the LXX—Their Nature and
Distribution,” in: D. Fraenkel and others (eds.), Studien zur Septuagin-
ta—Robert Hanhart zu Ehren (Mitteilungen der Septuaginta Unter-
nehmung XX; Gottingen 1990) 64-73.
18. “Theologically Motivated Exegesis Embedded in the Septua-
gint,” Proceedings of a Conference at the Annenberg Research Institute
May 15-16, 1989 (A JOR Supplement 1990; Philadelphia 1990) 215-233.
19. “The Textual Affiliations of 4QSam4,” JSOT 14 (1979) 37-53;
reprinted in E. Tov (ed.), The Hebrew and Greek Texts of Samuel, 1980
Proceedings IOSCS, Vienna (Academon: Jerusalem 1980) 189-205.
20. “The Contribution of the Qumran Scrolls to the Understanding of
the LXX,” in: G.J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and
Cognate Writings: Papers Presented to the International Symposium on
the Septuagint and Its Relations to the Dead Sea Scrolls and Other
Writings (Manchester, 1990) (SCS 33; Atlanta, GA 1992) 11-47.
21. “Interchanges of Consonants between the Masoretic Text and the
Vorlage of the Septuagint,” in: M. Fishbane and E. Tov (eds.),
“Shatarei Talmon”—Studies in the Bible, Qumran, and the Ancient
Near East Presented to Shemaryahu Talmon (Winona Lake, IN 1992)
295-266.
22. “Exegetical Notes on the Hebrew Vorlage of the LXX of Jeremiah
27 (34),” ZAW 91 (1979) 73-93.
23. “The Composition of 1 Samuel 17-18 in the Light of the Evidence
of the Septuagint Version,” in: J.H. Tigay (ed.), Empirical Models for
Biblical Criticism (Philadelphia 1985) 97-130.
24. “The Literary History of the Book of Jeremiah in the Light of Its
Textual History,” in: J.H. Tigay (ed.), Empirical Models for Biblical
Criticism (Philadelphia 1985) 211-237.
SOURCES XV
25. “The Growth of the Book of Joshua in the Light of the Evidence
of the LXX Translation,” Scripta Hierosolymitana 31 (1986) 321-339.
26. “Recensional Differences between the MT and LXX of Ezekiel,”
ETL 62 (1986) 89-101.
27. “Some Sequence Differences between the MT and LXX and Their
Ramifications for the Literary Criticism of the Bible,” JNSL 13 (1987)
151-160.
28. “Recensional Differences between the Masoretic Text and the
Septuagint of Proverbs,” in: H.W. Attridge and others (eds.), Of Scribes
and Scrolls, Studies on the Hebrew Bible, Intertestamental Judaism,
and Christian Origins Presented to John Strugnell (College Theology
Society Resources in Religion 5; Lanham, Maryland 1990) 43-56.
29. “Different Editions of the Song of Hannah,” in: M. Cogan, B.L.
Eichler, and J.H. Tigay (eds.), Tehillah le-Moshe, Biblical and Judaic
Studies in Honor of Moshe Greenberg (Winona Lake, IN 1997) 149-170.
30. “Pap. Giessen 13, 19, 22, 26: A Revision of the LXX?” RB 78 (1971)
355-383 and plates X—XI.
31. “Lucian and Proto-Lucian—Toward a New Solution of the
Problem,” RB 79 (1972) 101-113; reprinted in F.M. Cross and S. Talmon
(eds.), Qumran and the History of the Biblical Text (Cambridge, MA
1975) 293-305.
32. “The Methodology of Textual Criticism in Jewish Greek
Scriptures, with Special Attention to the Problems in Samuel—Kings—
The State of the Question: Problems and Proposed Solutions,” in: R.A.
Kraft (ed.), Septuagint and Cognate Studies 2 (1972) 3-15.
33. “Transliterations of Hebrew Words in the Greek Versions of the
Old Testament—A Further Characteristic of the kaige-Th. Revision?”
Textus 8 (1973) 78-92.
34. “Une inscription grecque d’origine samaritaine trouvée a
Thessalonique,” RB 81 (1974) 43-48.
35. “The Relation between the Greek Versions of Baruch and
Daniel,” in: M.E. Stone (ed.), Armenian and Biblical Studies (Jerusalem
1976) 27-34.
36. “The Textual History of the Song of Deborah in the A Text of the
LXX,” VT 28 (1978) 224-232.
37. “The ‘Lucianic’ Text of the Canonical and the Apocryphal
Sections of Esther: A Rewritten Biblical Book,” Textus 10 (1982) 1-25.
38. “The LXX Additions (Miscellanies) in 1 Kings 2,” Textus 11 (1984)
89-118.
LXX
LXXM2e)
LXX*
LxxXbLue
es
are
TJ
TN
EDITIONS OF TEXTUAL SOURCES
The individual volumes in the Gottingen Septuagint
series, when extant; otherwise the text of LXX is
quoted from the edition of Rahlfs, Septuaginta.
The individual volumes in the Gottingen Septuagint
series, when extant; otherwise the text of the
manuscript(s) is quoted from the editions of the
Cambridge series.
The ‘original’ text of the LXX reconstructed in the
Gottingen editions or in Rahlfs, Septuaginta as
opposed to later revisions correcting the translation
towards the proto-Masoretic text.
The Lucianic tradition (mainly manuscripts b,0,Cy,€5
according to the sigla used in the Cambridge
Septuagint) of the LXX, quoted according to the
Gottingen and Cambridge editions.
BHS
The Leiden edition of the Peshitta, when extant: The
Old Testament in Syriac According to the Peshitta
Version (Leiden 1966— ). Otherwise the edition of Lee
(London 1823) is quoted.
A. Tal, The Samaritan Pentateuch, Edited According
to MS 6 (C) of the Shekhem Synagogue (Texts and
Studies in the Hebrew Language and Related Subjects
8; Tel Aviv 1994)
M.L. Klein, The Fragment-Targums of the Pentateuch
According to their Extant Sources, vols. I-II (AnBib 76;
Rome 1980).
D. Rieder, Pseudo-Jonathan—Targum Jonathan ben
Uzziel on the Pentateuch Copied from the London MS
(Jerusalem 1974).
A. Diez Macho, Neophyti I, vols. I-V (Madrid/
Barcelona 1968-1978).
A. Sperber, The Bible in Aramaic Based on Old Manu-
scripts and Printed Texts, vols. I-I[Va (Leiden 1959-
1968).
XVI
EDITIONS
R. Weber, Biblia Sacra iuxta Vulgatam versionem (2d
ed.; Stuttgart 1975).
AASF
AB
AbrN
AJSL
AnBib
ANRW
AOAT
AOS
ASTI
ATAbh
BA
BASOR
BBLAK
BETL
Bib
BibOr
BiOr
BIOSCS
BJPES
BJRL
BK
BSac
BT
BWANT
BZ
BZAW
CATSS
CB
CBQ
CBQMS
PERIODICALS, REFERENCE WORKS, AND SERIALS
Annales Academiae Scientiarum Fennicae
Anchor Bible
Abr-Nahrain
American Journal of Semitic Languages and Literatures
Analecta Biblica
Aufstieg und Niedergang der rémischen Welt
Alter Orient und Altes Testament
American Oriental Series
Annual of the Swedish Theological Institute
Alttestamentliche Abhandlungen
Biblical Archaeologist
Bulletin of the American Schools of Oriental Research
Beitrage zur biblischen Landes- und Altertumskunde
Bibliotheca ephemeridum theologicarum lovaniensium
Biblica
Biblica et orientalia
Bibliotheca Orientalis
Bulletin of the International Organization for
Septuagint and Cognate Studies
Bulletin of the Jewish Palestine Exploration Society
Bulletin of the John Rylands University Library of
Manchester
Biblischer Kommentar
Bibliotheca Sacra
The Bible Translator
Beitrage zur Wissenschaft vom Alten und Neuen
Testament
Biblische Zeitschrift
Beihefte zur Zeitschrift fur die alttestamentliche
Wissenschaft
Computer Assisted Tools for Septuagint Studies
Cambridge Bible for Schools and Colleges
Catholic Biblical Quarterly
Catholic Biblical Quarterly Monograph Series
ConB
CR
DB
DBSup
EBib
EncBib
EncBrit
EncJud
Erlsr
EstBib
ETL
FRLANT
HAR
HAT
HSM
HSS
HThR
HUCA
ICC
IDB
IDBSup
LE}
IOMS
IOSCS
JANESCU
JAOS
JBL
JBR
JCS
JE
JJSt
JNESt
JNSL
JQR
JQRSup
ABBREVIATIONS
Coniectanea biblica
Classical Review
Dictionnaire de la Bible
Dictionnaire de la Bible, Supplément
Etudes bibliques
Encyclopaedia Biblica (Heb.)
Encyclopaedia Britannica
Encyclopaedia Judaica
Eretz Israel
Estudios biblicos
Ephemerides theologicae lovanienses
Forschungen zur Religion und Literatur des Alten und
Neuen Testaments
Hebrew Annual Review
Handbuch zum Alten Testament
Harvard Semitic Monographs
Harvard Semitic Studies
Harvard Theological Review
Hebrew Union College Annual
International Critical Commentary
The Interpreter’s Dictionary of the Bible
The Interpreter’s Dictionary of the Bible,
Supplementary Volume
Israel Exploration Journal
The International Organization for Masoretic Studies
International Organization for Septuagint and Cognate
Studies
Journal of the Ancient Near Eastern Society of Columbia
University
Journal of the American Oriental Society
Journal of Biblical Literature
Journal of Bible and Religion
Journal of Cuneiform Studies
The Jewish Encyclopedia
Journal of Jewish Studies
Journal of Near Eastern Studies
Journal of Northwest Semitic Languages
Jewish Quarterly Review
Jewish Quarterly Review Supplement
JSJ
JSOT
JSOTSup
JSSt
JThSt
KAT
KEH
KHAT
MGW]
MSU
NAWG
NCB
NKZ
NTT
OBO
OCD
OLZ
OTS
PAAJR
PSBA
RB
RE]
RevQ
RHR
SBL
SBLDS
SBLMasS
SBT
SCS
ScrHier
SEA
ABBREVIATIONS
Journal for the Study of Judaism in the Persian,
Hellenistic and Roman Period
Journal for the Study of the Old Testament
Journal for the Study of the Old Testament—
Supplement Series
Journal of Semitic Studies
Journal of Theological Studies
Kommentar zum Alten Testament (Leipzig 1913— ;
Gutersloh 1962-_ )
Kurzgefasstes exegetisches Handbuch zum Alten
Testament (Leipzig 1838—_)
Kurzer Hand-Commentar zum Alten Testament
(Freiburg /Tubingen 1897-1903)
Monatsschrift fiir Geschichte und Wissenschaft des
Judentums
Mitteilungen des Septuaginta-Unternehmens
Nachrichten der Akademie der Wissenschaften in
Gottingen
New Century Bible
Neue kirchliche Zeitschrift
Nederlands Theologisch Tijdschrift
Orbis biblicus et orientalis
Oxford Classical Dictionary
Orientalische Literaturzeitung
Oudtestamentische Studién
Proceedings of the American Academy of Jewish
Research
Proceedings of the Society of Biblical Archaeology
Revue biblique
Revue des études juives
Revue de Qumran
Revue de l'histoire des religions
Society of Biblical Literature
Society of Biblical Literature Dissertation Series
Society of Biblical Literature Masoretic Series
Studies in Biblical Theology
Septuagint and Cognate Studies
Scripta Hierosolymitana
Svensk Exegetisk Arsbok
ABBREVIATIONS
Scandinavian Journal of the Old Testament
The Society for Old Testament Study
Studies on the Texts of the Desert of Judah
Studia Theologica Lundensia
Studia in Veteris Testamenti Pseudepigrapha
Theologische Literaturzeitung
Theologische Realenzyklopadie
Theologische Rundschau
Theologische Quartalschrift
Texts and Studies
Texte und Studien zum Antiken Judentum
Theologische Studien und Kritiken
Texte und Untersuchungen
Tyndale Bulletin
Ugarit-Forschungen
Vetus Testamentum
Vetus Testamentum, Supplements
Die Welt des Orients
Westminster Theological Journal
Wissenschaftliche Monographien zum Alten und Neuen
Testament
Wissenschaftliche Untersuchungen zum Neuen
Testament
Zeitschrift fiir die alttestamentliche Wissenschaft
Zeitschrift der deutschen morgenlandischen
Gesellschaft
BIBLIOGRAPHICAL ABBREVIATIONS
Actes 1986
Actes du Premier Colloque Internationale Bible et Informatique: Le texte,
Louvain-la-Neuve (Belgique) 2-3-4 septembre 1985 (Paris/Genéve 1986)
Allen, Chronicles
L.C. Allen, The Greek Chronicles, 1 (VTSup 25; Leiden 1974)
Aptowitzer, “Berichte”
V. Aptowitzer, “Die rabbinischen Berichte uber die Entstehung der
Septuaginta,” Hagedem 2 (1909) 11-27, 102-122; 3 (1910) 4-17
BAGD
W.F. Arndt, F.W. Gingrich, and F.W. Danker, A Greek-English Lexicon of
the New Testament and Other Early Christian Literature (2d ed.;
Chicago/London 1979)
Barr, Comparative Philology
J. Barr, Comparative Philology and the Text of the Old Testament (Oxford
1986; repr. Winona Lake, IN 1987)
Barthélemy, Devanciers
D. Barthélemy, Les devanciers d’Aquila (VTSup 10; Leiden 1963)
—, Etudes
—, Etudes d'histoire du texte de l'Ancien Testament (OBO 21; Fribourg/
Gottingen 1978)
—,CT I
—, Critique textuelle de l’Ancien Testament, 1 (OBO 50/1; Fribourg/
Gottingen 1982)
Bauer, Worterbuch
W. Bauer, Griechisch-Deutsches Worterbuch zu den Schriften des Neuen
Testaments (5th ed.; Berlin 1958)
XXIV ABBREVIATIONS
BDB
S.R. Driver, F. Brown, and C.A. Briggs, A Hebrew and English Lexicon of the
Old Testament (Oxford 1957)
BH
Biblia Hebraica, Edited by R. Kittel and P. Kahle (3rd [7th] ed.; Stuttgart
1951)
BHS
Biblia Hebraica Stuttgartensia, Edited by K. Elliger and W. Rudolph
(Stuttgart 1976-1977)
Bible grecque
M. Harl, G. Dorival, and O. Munnich, La Bible grecque des Septante—Du
judaisme hellénistique au christianisme ancten (Paris 1988)
Bogaert, Le livre de Jérémie
P.-M. Bogaert, Le livre de Jérémie. Le prophéte et son milieu, les oracles et
leur transmission (BETL 54; Leuven 1981)
—, “De Baruch a Jérémie”
—, “De Baruch a Jérémie, les deux rédactions conservées du livre de
Jérémie” in: Bogaert, Le livre de Jérémie, 168-173
—,“Mécanismes”
—, “Les mécanismes rédactionnels en Jér 10, 1-16 (LXX et TM) et la
signification des suppléments” in: Bogaert, Le livre de Jérémie, 222-238
Brooke—McLean
A.E. Brooke, N. McLean, and H.St.J. Thackeray, The Old Testament in
Greek According to the Text of Codex Vaticanus (Cambridge 1906-1940)
Brooke—Garcia Martinez, New Qumran Texts
G.J. Brooke with F. Garcia Martinez (eds.), New Qumran Texts and
Studies—Proceedings of the First Meeting of the International Organization
for Qumran Studies, Paris 1992 (STDJ XV, Leiden/New York/Koln 1994)
Camilo dos Santos, Index
E. Camilo dos Santos, An Expanded Hebrew Index for the Hatch-Redpath
Concordance to the Septuagint (Jerusalem [1973]})
CATSS 1
J.R. Abercrombie, W. Adler, R.A. Kraft, and E. Tov: Computer Assisted
Tools for Septuagint Studies (CATSS), Volume 1, Ruth (SCS 20; Atlanta, GA
1986)
ABBREVIATIONS XXV
CATSS 2
E. Tov, A Computerized Data Base for Septuagint Studies—The Parallel
Aligned Text of the Greek and Hebrew Bible, CATSS Volume 2 (JNSL,
Supplementary Series 1; 1986)
CAITSS 3
J. Jarick with the collaboration of G. Marquis, A Comprehensive Bilingual
Concordance of the Hebrew and Greek Texts of the Book of Ecclesiastes
(CATSS—Basic Tools Volume 3; SCS 36; Atlanta, GA 1993)
Classified Bibliography
S.P. Brock et al., A Classified Bibliography of the Septuagint (Arbeiten zur
Literatur und Geschichte des hellenistischen Judentums 6; Leiden 1973)
Cook, “A Text”
H.J. Cook, “The A Text of the Greek Versions of the Book of Esther,” ZAW
81 (1969) 369-376
Cox, VI Congress
C.E. Cox (ed.), VI Congress of the International Organization for Septuagint
and Cognate Studies Jerusalem 1986 (SCS 23; Atlanta GA 1987)
—, VII Congress
—, VII Congress of the International Organization for Septuagint and
Cognate Studies Leuven 1989 (SCS 31; Atlanta, GA 1991)
Cross, “New Qumran Fragment”
F.M. Cross, “A New Qumran Fragment Relating to the Original Hebrew
Underlying the Septuagint,” BASOR 132 (1953) 15-26
—, “Biblical Text”
—, “The History of the Biblical Text in the Light of Discoveries in the
Judaean Desert”, HThR 57 (1964) 281-299
—, “Evolution”
—, “The Evolution of a Theory of Local Texts,” in: Cross-Talmon, QHBT,
306-320
— ALQ3
—, The Ancient Library of Qumran (3d ed.; Sheffield 1995)
Cross—Talmon, QHBT
F.M. Cross and S. Talmon (eds.), Qumran and the History of the Biblical Text
(Cambridge, MA/London 1976)
XXVI ABBREVIATIONS
Daniel, Recherches
S. Daniel, Recherches sur le vocabulaire du culte dans la Septante (Paris
1966)
Delitzsch, Lese- und Schreibfehler
F. Delitzsch, Die Lese- und Schretbfehler im Alten Testament nebst den dem
Schrifttexte einverleibten Randnoten Klassifiziert (Berlin/Leipzig 1920)
Dijkstra, “Glosses”
M. Dijkstra, ‘The Glosses in Ezekiel Reconsidered: Aspects of Textual
Transmission in Ezekiel 10,”” in: Lust, Ezekiel, 55-77
DJD
Discoveries in the Judaean Desert (of Jordan), vols. I~ (Oxford 1955— )
DJD Il
M. Baillet and others, Les ‘petites grottes’ de Qumran (DJD III; Oxford
1962)
DID V
J.M. Allegro, Qumran Cave 4.1 (40158-4Q186) (DJD V; Oxford 1968)
DJD VIII
E. Tov with the collaboration of R.A. Kraft, The Greek Minor Prophets
Scroll from Nahal Hever (8HevXIlgr) (The Setyal Collection I) (DJD VIII;
Oxford 1990)
DJD IX
P.W. Skehan, E. Ulrich, and J.E. Sanderson, Qumran Cave 4.1V, Palaeo-
Hebrew and Greek Biblical Manuscripts (DJD IX; Oxford 1992)
DJD xi
E. Ulrich and F.M. Cross (eds.), Qumran Cave 4.VII: Genesis to Numbers
(DJD XI]; Oxford 1994)
DJD XIV
E. Ulrich and F.M. Cross (eds.), Qumran Cave 4.1X: Deuteronomy, Joshua,
Judges, Kings (DJD XIV; Oxford 1995)
DJD XV
E. Ulrich and others, Qumran Cave 4.X: The Prophets (DJD XV; Oxford
1997)
Dogniez, Bibliography
C. Dogniez, Bibliography of the Septuagint, Bibliographie de la Septante
1970-1993 (VTSup 60; Leiden 1995)
ABBREVIATIONS XXVII
Driver, Samuel
S.R. Driver, Notes on the Hebrew Text and the Topography of the Books of
Samuel, with an Introduction on Hebrew Palaeography and the Ancient
Versions (2d ed.; Oxford 1913)
—, Introduction
—, An Introduction to the Literature of the Old Testament (9th ed.;
Edinburgh 1913)
Driver, ‘“Glosses”
G.R. Driver, ‘’“Glosses in the Hebrew Text of the Old Testament,” l’Ancien
Testament et l’Orient (Orientalia et Biblica Lovaniensia 1; Louvain 1957)
123-161
Eissfeldt, Introduction
O. Eissfeldt, The Old Testament, An Introduction, Including the Apocrypha
and Pseudepigrapha, and also the Works of Similar Type from Qumran. The
History of the Formation of the Old Testament (trans. P.R. Ackroyd; Oxford
1965)
Fernandez Marcos, Introduccion
N. Fernandez Marcos, Introduccion a las versiones griegas de la Biblia
(Textos y Estudios “Cardenal Cisneros” 23; Madrid 1979)
—, La Septuaginta
N. Fernandez Marcos (ed.), La Septuaginta en la investigacion
contemporanea (V Congreso de la IOSCS) (Textos y Estudios “Cardenal
Cisneros” 34; Madrid 1985)
—, Scribes
—, Scribes and Translators—Septuagint and Old Latin in the Books of Kings
(VTSup 54; Leiden 1994)
Field, Hexapl.
F, Field, Origenis Hexaplorum quae supersunt (Oxford 1875)
Fohrer, ““Ezechiel”
G. Fohrer, ‘’Die Glossen im Buche Ezechiel,’” ZAW 63 (1951) 33-53 =
BZAW 99 (1967) 204-221
Fraenkel, Studien zur Septuaginta
D. Fraenkel and others (eds.), Studien zur Septuaginta—Robert Hanhart zu
Ehren (MSU XX; Gottingen, 1990)
Frankel, Vorstudien
Z. Frankel, Vorstudien zu der Septuaginta (Leipzig 1841)
XXVIII ABBREVIATIONS
Frankel, Einfluss
—, Uber den Einfluss der palastinischen Exegese auf die alexandrinische
Hermeneutik (Leipzig 1851)
Freedy, ‘’Ezekiel”
K.S. Freedy, “The Glosses in Ezekiel I-XXIV,’”’ VT 20 (1970) 129-152
Friedmann, Onkelos
M. Friedmann, Onkelos und Akylas (Wien 1896)
Geiger, Urschrift
A. Geiger, Urschrift und Ubersetzungen der Bibel (Breslau 1857)
Gerleman, Job
G. Gerleman, Studies in the Septuagint, I. Book of Job (LUA NEF 43, 2; Lund
1946)
Gesenius, Pent. sam.
W. Gesenius, De Pentateuchi samaritani origine, indole et auctoritate
commentatio philologico-critica (Halle 1815)
Gesenius—Kautzsch, Grammar
E. Kautzsch, Gesenius’ Hebrew Grammar (2d ed.; Oxford 1910)
Goldman, Prophétie
Y. Goldman, Prophétie et royauté au retour de l’exil—Les origines littéraires
de la forme massorétique du livre de Jérémie (OBO 118; Freiburg /Gottingen
1992)
Gooding, “Text and Midrash”
D.W. Godding , “Problems of Text and Midrash in the Third Book of
Reigns,” Textus 7 (1969) 1-29
—, Relics
D.W. Gooding, Relics of Ancient Exegesis, A Study of the Miscellanies in 3
Reigns 2 (SOTS, Monograph Series 4, Cambridge 1976)
Greenspoon—Munnich, VIII Congress
L. Greenspoon and O. Munnich, VIII Congress of the International
Organization for Septuagint and Cognate Studies, Paris 1992 (SCS 41;
Atlanta, GA 1995)
Hall, Companion
F.W. Hall, A Companion to Classical Texts (Oxford 1913; repr. Chicago
1970)
Hanhart, Esther
R. Hanhart, Septuaginta, Vetus Testamentum graecum in auctoritate
academiae litterarum gottingensis editum, VIII, 3 (Gottingen 1966)
ABBREVIATIONS XXIX
Hatch, Essays
E. Hatch, Essays in Biblical Greek (Oxford 1889)
Helbing, Grammatik
R. Helbing, Grammatik der Septuaginta, Laut- und Wortlehre (Gottingen
1907)
Higger, Soferim
M. Higger, 077910 nd0n (New York 1937)
HR
E. Hatch and H.A. Redpath, A Concordance to the Septuagint and the Other
Greek Versions of the Old Testament (Including the Apocryphal Books), I-III
(Oxford, 1892-1906; repr. Graz 1954; 2d ed.; Grand Rapids, MI 1998
[with a Hebrew/ Aramaic Index by T. Muraoka])
Janzen, Jeremiah
J.G. Janzen, Studies in the Text of Jeremiah (HSM 6; Cambridge, MA 1973)
Jellicoe, SMS
S. Jellicoe, The Septuagint and Modern Study (Oxford 1968)
Johannessohn, Prdpositionen
M. Johannessohn, Der Gebrauch der Kasus und der Pripositionen in der
Septuaginta, I: Gebrauch der Kasus (Berlin 1910)
Johnson, Rezension
B. Johnson, Die hexaplarische Rezension des 1 Samuelbuches der Septua-
ginta (STL 22; Lund 1963)
Kahle, Cairo Geniza
P. Kahle, The Catro Geniza (2d ed.; Oxford 1959)
Kraft, Lexicography
R.A. Kraft (ed.), Septuagintal Lexicography (SCS 1; Missoula, Montana
1972)
Kutscher, Language
Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll
(1QIs7) (Leiden 1974)
de Lagarde, Proverbien
P.A. de Lagarde, Anmerkungen zumr griechischen Ubersetzung der
Proverbien (Leipzig 1863)
Langen, “Esther”
J. Langen, “Die beiden griechischen Texte des Buches Esther,” TQ 42 (1860)
244-272
XXX ABBREVIATIONS
Lee, Lexical Study
J.A.L. Lee, A Lexical Study of the Septuagint Verion of the Pentateuch (SCS
14; Chico, CA 1983)
Levine, Aramaic Version
E. Levine, The Aramaic Version of the Bible: Contents and Context (BZAW
174; 1988)
Lewis, “Hannah”
Th.J. Lewis, “The Textual History of the Song of Hannah: 1 Samuel IT 1-
10,” VT 44 (1994) 18-46
LSJ
H.G. Liddell, R. Scott and H.S. Jones, A Greek-English Lexicon, (9th ed.;
Oxford, 1940); LSJ is used together with LSJ, Supplement and P.G.W. Glare,
Revised Supplement (Oxford 1996)
LSJ, Supplement
E.A. Barber, A Greek-English Lexicon, A Supplement (Oxford 1968)
Lust, Ezekiel
J. Lust (ed.), Ezekiel and his Book, Textual and Literary Criticism and Their
Interrelation (BETL 74; Leuven 1986)
—, “Vocabulary”
—, “The Vocabulary of LXX Ezekiel and its Dependence upon the
Pentateuch,” in: M. Vervenne and J. Lust (eds.), Deuteronomy and
Deuteronomic Literature, Festschrift C.H.W. Brekelmans (BETL 123; Leuven
1997) 529-546
Maas, Textual Criticism
P. Maas, Textual Criticism (trans. B. Flower; Oxford 1958) = Textkritik, in:
A. Gercke and E. Norden, Einleitung in die Altertumswissenschaft, I, VII
(3d ed.; Leipzig 1957)
Manchester Symposium
G. J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and Cognate Writings:
Papers Presented to the International Symposium on the Septuagint and Its
Relations to the Dead Sea Scrolls and Other Writings (Manchester, 1990)
(SCS 33 ; Atlanta, GA 1992)
Margolis, “Studien”
M.L. Margolis, “Studien im griechischen Alten Testament,” ZAW 27 (1907)
212-270
ABBREVIATIONS XXXI
Margolis, “Complete Induction”
—, “Complete Induction for the Identification of the Vocabulary in the
Greek Versions of the Old Testament with Its Semitic Equivalents—Its
Necessity and the Means of Obtaining It,” JAOS 30 (1910) 301-312
— “Scope”
—, “The Scope and Methodology of Biblical Philology,” JQR 1 (1910-
1911) 5-41
—, “Specimen”
—, “Specimen of a New Edition of the Greek Joshua,” Jewish Studies in
Memory of Israel Abrahams (New York 1927; repr. 1980) 307-323
—, Joshua
—, The Book of Joshua 1n Greek According to the Critically Restored Text
with an Apparatus Containing the Variants of the Principal Recensions and
of the Individual Witnesses, I-IV (Paris 1931 [- 1938]), V (Philadelphia
1992)
McCarter, Samuel
P.K. McCarter, I Samuel, A New Translation with Introduction and
Commentary (AB; Garden City, NY 1980).
—, Textual Criticism
—, Textual Criticism. Recovering the Text of the Hebrew Bible (Philadelphia
1986)
Min, Minuses and Pluses
Y.-J. Min, The Minuses and Pluses of the LXX Translation of Jeremiah as
Compared with the Massoretic Text: Their Classification and Possible
Origins, unpubl. diss. Hebrew University (Jerusalem 1977)
MM
J.P.H. Moulton and G. Milligan, The Vocabulary of the Greek Testament
Illustrated from the Papyri and Other Non-Literary Sources (London 1930)
Moore, “Greek Witness”
C.A. Moore, “A Greek Witness to a Different Hebrew Text of Esther,”
ZAW 79 (1967) 351-358
—, Esther
—, Esther (AB; Garden City, New York 1971)
—, Additions
—, Daniel, Esther and Jeremiah, The Additions (AB; Garden City, New York
1977)
XXXII ABBREVIATIONS
Muller, “Nachrichten”
K. Muller, “Die rabbinischen Nachrichten tiber die Anfange der
Septuaginta,” Forschung zur Bibel, Festschrift J. Ziegler (Wurzburg 1972)
73-93
Muraoka, “Septuagint Lexicon”
T. Muraoka, “Towards a Septuagint Lexicon,” in: Cox, VI Congress, 255—
276
—, Melbourne Symposium
— (ed.), Melbourne Symposium on Septuagint Lexicography (SCS 28;
Atlanta, GA 1990)
NAB
The New American Bible (New York/London 1970)
NEB
The New English Bible with the Apocrypha (Oxford/Cambridge 1970)
NJPST
Tanakh, The Holy Scriptures, The New JPS Translation According to the
Traditional Hebrew Text (The Jewish Publication Society; Philadelphia/
New York/Jerusalem 1988)
NRSV
The Holy Bible Containing the Old and New Testaments with the
Apocryphal/Deuterocanonical Books, New Revised Standard Version
(Glasgow /London 1989)
O’Connell, Exodus
K.G. O’Connell, The Theodotionic Revision of the Book of Exodus (HSM 3;
Cambridge, MA 1972)
Peters, Beitrage
Peters, Beitrage zur Text- und Literarkritik sowie zur Erklarung der Bticher
Samuel (Freiburg i. Breisgau 1899)
Pietersma—Cox, De Septuaginta
A. Pietersma and C.E. Cox (eds.), De Septuaginta, Studies in Honour of J].W.
Wevers on His Sixty-Fifth Birthday (Mississauga, Ont. 1984)
Rahlfs, Septuaginta
A. Rahlfs, Septuaginta, id est Vetus Testamentum graece iuxta LXX
interpretes (Stuttgart 1935)
ABBREVIATIONS XXXII
Rahlfs, Lucian’s Rezension
—-, Lucian’s Rezension der Konigsbticher (Septuaginta Studien 3; Géttingen
1911)
REB
The Revised English Bible with the Apocrypha (Oxford/Cambridge 1989)
Reider, Prolegomena
J. Reider, Prolegomena to a Greek-Hebrew and Hebrew-Greek Index to Aquila
(Philadelphia 1916)
Roberts, Text
B.J. Roberts, The Old Testament Text and Versions—The Hebrew Text in
Transmission and the History of the Ancient Versions (Cardiff 1951)
RSV
The Bible, Containing the Old and New Testaments, Revised Standard
Version (The British and Foreign Bible Society, 1971)
Salvesen, Origen’s Hexapla
A. Salvesen (ed.), Origen’s Hexapla and Fragments (TSAJ 58; Tubingen
1998)
Schleusner, Thesaurus
J.F. Schleusner, Novus thesaurus philologico-criticus stve lexicon in LXX et
reliquos interpretes graecos ac scriptores apocryphos Veteris Testamenti
(Leipzig 1820-1821; Glasgow 1822; London 1829)
Schmitt, Theodotion
A. Schmitt, Stammt der sogenannte “8
Theodotion? MSU IX (Gottingen 1966)
ta
Text bet Daniel wirklich von
Second Colloquium
Proceedings of the Second International Colloquium Bible and Computer:
Methods, Tools, Results, Jérusalem, 9-13 Juin 1988 (Paris/Genéve 1989)
Seeligmann, JEOL
I.L. Seeligmann, “Problemen en perspectieven in het moderne Septuaginta
Onderzoek,” Jaarbericht van het Vooraziatisch-Egyptisch Gezelschap “Ex
Oriente Lux” 7 (1940) 359-390e
—, Isaiah
—, The Septuagint Version of Isaiah (Leiden, 1948)
—, “Problems”
—,”Problems and Perspectives in Modern Septuagint Research,” Textus 15
(1990) 169-232 [translation of Seeligmann, JEOL]
XXXIV ABBREVIATIONS
Shenkel, Chronology
J.D. Shenkel, Chronology and Textual Development in the Greek Text of
Kings (HSM 1; Cambridge, MA 1968)
Smith, Samuel
H.P. Smith, The Books of Samuel (ICC; Edinburgh 1899)
Sollamo, Semtprepositions
R. Sollamo, Renderings of Hebrew Semiprepositions in the Septuagint
(AASF, Diss. Hum. Litt. 19; Helsinki 1979)
Sperber, Grammar
A. Sperber, A Historical Grammar of Biblical Hebrew—A Presentation of
Problems with Suggestions to Their Solution (Leiden 1966)
Stoebe, Samuel
H.J. Stoebe, Das erste Buch Samuelis (KAT; Gutersloh 1973) 100-107
Swete, Introduction
H.B. Swete, An Introduction to the Old Testament in Greek (2d ed.;
Cambridge 1914)
Talmon, “Double Readings”
S. Talmon, “Double Readings in the Massoretic Text,” Textus 1 (1960)
144-184
—, “Scrolls”
—, The Three Scrolls of the Law that were found in the Temple Court,”
Textus 2 (1962) 14-27
—, “Old Testament Text”
—, “The Old Testament Text,” in: R.P. Ackroyd and C.F. Evans (eds.), The
Cambridge History of the Bible, vol. I (Cambridge, 1970) 159-199; repr. in
Cross—Talmon, QHBT (1976) 1-41
—, “Textual Study”
—, “The Textual Study of the Bible—A New Outlook,” in: Cross~Talmon,
QHBT 321-400
Taylor, 1X Congress
B.A. Taylor, IX Congress of the International Organization for Septuagint
and Cognate Studies, Cambridge, 1995 (SCS 45; Atlanta, GA 1997)
TCHB
see Tov, TCHB
TCU
see Tov, TCU
ABBREVIATIONS XXXV
Thackeray, Grammar
H.St.J. Thackeray, A Grammar of the Old Testament in Greek According to
the Septuagint (Cambridge 1909)
ThDNT
G. Kittel (ed.), Theological Dictionary of the New Testament, (trans. G.W.
Bromiley; Grand Rapids, MI 1968-1976)
Tigay, Models
J.H. Tigay (ed.), Empirical Models for Biblical Criticism (Philadelphia 1985)
Tournay, “Cantique”
R. Tournay, “Le Cantique d’Anne | Samuel II.1-10,” in: Mélanges
Dominique Barthélemy. Etudes bibliques offertes a l'occasion de son 60¢e
anniversaire (OBO 38; Fribourg/Géttingen 1981) 553-576
Tov, “L’incidence”
E. Tov, “L’incidence de la critique textuelle sur la critique littéraire dans le
livre de Jérémie,” RB 79 (1972) 189-199
—, Baruch
—, The Book of Baruch Also Called I Baruch (Greek and Hebrew) (Texts and
Translations 8, Pseudepigrapha Series 6; Missoula, MT 1975)
—, Jeremiah and Baruch
—, The Septuagint Translation of Jeremiah and Baruch—A Discussion of an
Early Revision of Jeremiah 29-52 and Baruch 1:1-3:8 (HSM 8; Missoula,
MT 1976)
—, “Compendia”
—, “The Septuagint,” in: Mulder, Mikra, 161-188
—, TCHB
—, Textual Criticism of the Hebrew Bible (Minneapolis and Assen/
Maastricht, 1992)
—, TCU
—, The Text-Critical Use of the Septuagint in Biblical Research (2d ed.;
Jerusalem Biblical Studies 8; Jerusalem 1997)
Tov: References with a star refer to chapters in this book
—, “Baruch”*: Chapter 35
—, “Compound Words”*: Chapter 9
—, “Computerized Database”*: Chapter 3
—, “Deborah”*: Chapter 36
XXXVI ABBREVIATIONS
Tov, “Dimensions”*: Chapter 6
—, “Esther’*: Chapter 37
—, “Exegetical Notes”*: Chapter 22
—, “Ezekiel”*: Chapter 26
—, “Glosses”*: Chapter 4
—, “Greek Words”*: Chapter 8
—, “Hiphal’*: Chapter 13
—, “Infinitive Absolute”*: Chapter 17
—, “Inscription samaritaine”*: Chapter 34
—, “Interchanges”*: Chapter 21
—, “Jeremiah’”*: Chapter 24
—, “Joshua”*: Chapter 25
—, “Lexicon”*: Chapter 7
—, “Loan-words”*: Chapter 11
—, “Lucian”*: Chapter 31
—, “Margolis”*: Chapter 2
—, “Methodology”*: Chapter 32
—, “Midrash Joshua”*: Chapter 10
—, “Miscellanies’*: Chapter 38
—, “Pap. Giessen”*: Chapter 30
—, “Proverbs”*: Chapter 28
—, “Qumran’*: Chapter 20
—, “Rabbinic Tradition”*: Chapter 1
—, “Review Veltry”*: Chapter 5
—, “Samuel”’*: Chapter 23
—, “Sequence Differences”*: Chapter 27
—, “Song of Hannah”*: Chapter 29
—, “Theology’”*: Chapter 18
—, “Torah’*: Chapter 12
—, “Translation Technique”*: Chapter 16
—, “Transliterations”*: Chapter 33
—, “Understand”*: Chapter 14
—, “4QSam?”*: Chapter 19
Tov-Wright, “Literalness”*: Chapter 15
Trebolle, Madrid Qumran Congress
J. Trebolle Barrera and L. Vegas Montaner (eds.), The Madrid Qumran
Congress—Proceedings of the International Congress on the Dead Sea
Scrolls—Madrid, 18-21 March, 1991 (STDJ 11; Leiden/Madrid 1992)
ABBREVIATIONS XXXVII
Ulrich, Samuel
E. Ulrich, The Qumran Text of Samuel and Josephus (HSM 19; Missoula,
MT 1978)
Walters, Text
P. Walters, The Text of the Septuagint, Its Corruptions and Their
Emendation (Cambridge 1973)
Walters, “Hannah and Anna”
S.D. Walters, “Hannah and Anna: The Greek and Hebrew Texts of 1
Samuel 1,” JBL 107 (1988) 385-412
Weinfeld, Deuteronomy
M. Weinfeld, Deuteronomy and the Deuteronomic School (Oxford 1972)
Wellhausen, Samuel
J. Wellhausen, Der Text der Biicher Samuelis (GOttingen 1871)
Wutz, Transkriptionen
F.X. Wutz, Die Transkriptionen von der Septuaginta bis zu Hieronymus
(Stuttgart 1925-1933)
—, Systematische Wege
—, Systematische Wege von der Septuaginta zum hebraischen Urtext
(Stuttgart 1937)
Ziegler, Untersuchungen
J. Ziegler, Untersuchungen zur Septuaginta des Buches Isaias, ATA XII, 3
(1934)
—, Beitrage
—, Beitrage zur leremtas-Septuaginta, NAWG, Phil-hist. Kl]. 1958, 2
(Gottingen, 1958)
—, leremias
—, Ieremias. Baruch. Threni. Epistula leremiae, Septuaginta, Vetus Testa-
mentum graecum etc. XV (2d ed.; Gottingen 1976)
XXX VIII
ABBREVIATIONS
OTHER ABBREVIATIONS
Aquila
Babylonian Talmud
The Damascus Document from the Cairo Geniza
Septuagint (= Greek)
Hieronymus, Jerome
Jerusalem Talmud
Ketiv
Vetus Latina (Old Latin)
Septuagint
Masoretic Text
Old Greek (translation of the LXX)
Qere
Peshitta
Samaritan Pentateuch
Targum of the Samaritan Pentateuch
Symmachus
Targum
Targum Jonathan
Targum Ongelos
Targum Pseudo-Jonathan
Theodotion
Vulgate
I. GENERAL STUDIES
CHAPTER ONE
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’
INSERTED INTO THE GREEK TRANSLATION OF THE TORAH
AND THEIR RELATION TO THE ORIGINAL TEXT
OF THE SEPTUAGINT
Various passages within rabbinic literature cite a series of alterations
which were inserted into the Greek translation of the Torah. In these
passages a list of 10 (11), 13, 15, or 18 (16) such alterations appears
along with a brief account of the circumstances under which they were
inserted in the translation. The background of this rabbinic tradition is
examined here, as well as its importance for LXX studies. Special
attention is given to the implications of the exact wording of the list for
our understanding of the original form of the LXX.
1. The sources
The principal sources for the rabbinic tradition are: b. Meg. 9a; y. Meg.
1, 1, 4., p. 72a; Mek. Exod 12, 40; Midr. Hagadol Exod 4, 20; Abot de-R.
Nat. version B, chapter 37; Soph. 1. 7; Yal. Shim. Gen 3; Midr. Tan.
Exod para 22. Additional sources are listed in Higger, Soferim, 101.
2. The list
The various sources list a different number of alterations and at times
explicitly state the number at the head of the list. Thus Abot de-R.
Nat. and Midr. Tan. Exod paragraph 22 mentions 10 alterations (al-
though the lists include 11 or 14 instances) and Midr. Hagadol on Exod
4:20 and Deut 4:19 mentions 18 alterations (the list in Exodus includes
only 16 alterations). Other lists do not indicate any number at the head
of their lists: b. Meg. 9a; Mek. Exod 12:40; Yal. Shim. Gen, paragraph 3.
It would be natural to assume that the shortest list (10 or 11
alterations) reflects the original formulation of the rabbinic tradition,
2 CHAPTER ONE
expanded by the longer lists; however, the list and the story associated
with it developed not only by expansion but also by abridgment.
The sources mentioning 13 or 15 alterations are the most widespread
and presumably reflect the central tradition. The difference between
these two traditions lies in the inclusion or exclusion of passages 10 and
11. Attention should be drawn to the fact that the list with 16
alterations (Midr. Hagadol Exod 4:20) came about as a result of the
addition of biblical passages similar to those originally in the list, and
that list is therefore secondary. Among the other traditions, 10 or 18
alterations are mentioned in the headings of the list (though the lists
themselves contain some other figure); it would appear that these
figures have been influenced by other lists of 10 items in the context
(Abot de-R. Nat. ibid.; Abot chapter 5, 1-9) and in the same way by the
list of 18 emendations of the scribes in the Hebrew text of the Bible,
which, too, is known from rabbinic literature.! Tendencies toward
expansion and abridgment are also noticeable in the items comprising
the list themselves, both regarding the biblical citations and their
explanations. This problem is particularly acute in light of the fact
that certain citations reflect more than one alteration (see notes 28, 29).
In view of these considerations it is impossible to determine with
certainty which among the above-mentioned lists is the original or the
nearest to it. The lists in b. Meg., y. Meg. and Mek. are the most ancient
among the sources, but we lack proven criteria in order to evaluate the
differences between these sources themselves. Furthermore, each list
itself is transmitted in various forms, both in manuscripts and printed
editions, so it is hard to determine their original form, if that existed
at all. There were also mutual influences between the various lists, at
least at the level of individual manuscripts.
The relationship between the different sources was described in
general terms by Frankel, Friedmann, Geiger, Aptowitzer, and Miiller.*
Before Aptowitzer it was generally believed that the relatively short
baraita (13 passages) in y. Meg. (and similarly the list in Mek.) reflects
a more original form than the other sources, but Aptowitzer considered
the baraita in b. Meg. earlier. These two opinions are supported by
different arguments (see Aptowitzer, “Berichte” 3 [1910] 102 ff.);
1 See Mek. Exod 15:7, Sifre Num 10:35 ef al. For an analysis, see Geiger, Urschrift, 231-261;
B. Keller, “Fragment d’un traité d’exégése massorétique,” Textus 5 (1966) 60-84; W.E.
Barnes, “Ancient Corrections in the Text of the O.T.,” JTS 1 (1900) 379-414; W. McKane,
“Observations on the Tikkiiné Sép*rim,” Festschrift Eugene A. Nida (The Hague/Paris 1974)
53-77.
2 Frankel, Vorstudien; Friedmann, Onkelos; Geiger, Urtext; Aptowitzer, “Berichte”; Miiller,
“Nachrichten.”
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 3
evidently the main problem is the inclusion or exclusion of passages 10
and 11. In b. Meg. these passages are included in the list, while in y.
Meg. and in Mek. they are lacking. Judging by their contents, these
passages belong in the list, but it is hard to determine if they also
appeared at the earliest stage of its development.’ Even if these
passages were added to the list only at a later stage, the discussion
will turn out to be profitable if it is based on the longest of the ancient
lists. To that end, the 15 passages included in the list of b. Meg. 9a are
cited below according to their sequence in the Talmud, quoted from the
Vilna edition, and accompanied by variants from MS Munchen (quoted
from R. Rabinowitz, 0°9510 >717}27, 8 [Miinchen 1877]) and other sources.*
y. Meg. 1,1, 4., p. 71b
Mek. Exod 12:40 according to H.S. Horowitz-Rabin (2d ed.; Jerusalem
1960)
Midr. Hagadol Exod 12:40 according to M. Margoliouth (Jerusalem
1967)
Abot de-R. Nat., version B, chapter 37 according to S. Schechter
(Vienna 1887)
Soph. 1.7 according to Higger, Soferim; individual manuscripts are
here quoted as ‘Soph., mss’
Yal. Shim. Gen, paragraph 3 according to the edition of the Rav
Kook Institute (Jerusalem 1973)
Midr. Tan. Exod paragraph 22
1. MPwRID NID AT?N (Gen 1:1)
2. 0092) 0982 OTN WYK (Gen 1:26)
miniai] Abot de-R. Nat.: mint). Soph. pr.: o7°98 Tox. Midr. Hagadol
Exod adds: n721 0742 OFN7 nx o>R NII (Gen 1:27).
3. (yawn ova maw ww ora 23 (Gen 2:2)
y. Meg. and Soph.: *y"1w2 n(1)aw™ ~wwa 22>. In most traditions (except
for Midr. Hagadol Exod) 0°72 of MT is lacking. In y. Meg., Mek., Midr.
3 Regarding this detail, is the list of the Yerushalmi earlier since the problematic passages
are not found there; or perhaps were they omitted from the list in the Yerushalmi because
they were problematic? Similarly, passage 15 appears in its present place in b. Meg. out of
the verse order and should thus be considered an addition. On the other hand, it appears
in y. Meg. in its proper place according to the order of the passages. It is hard to determine
whether it was inserted here later or whether this was its original place.
A perusal of the various manuscripts of these sources reveals that the many variant
readings listed below as variants between the different lists appear also as variants within
the tradition of b. Meg. (and also in other traditions, e.g., Higger, Soferim). The manuscripts
of b. Meg. are not listed below. For example, if for passage 4, it is written according to our
principles, that the words OX72 13N> XN?) are lacking in manuscript M of b. Meg. it should be
pointed out that they are actually lacking in all the major manuscripts.
4 CHAPTER ONE
Hagadol Exod and Tan., passages 3 and 4 are cited in inverted order. In
Abot de-R. Nat. this passage is lacking.
4. ON72 13ND X77 NID AAPN 7ADr (Gen 5:2)
napijy. Meg.: rap; Mek., Midr. Hagadol Exod, Yal. Shim. and
Soph.: raipn (thus also Gen Rab. 8:11).
wa] Mek. and Soph.: ox 13; Midr. Hagadol Exod: 1nX x72 (the full
text 1S: INT2 PIN ADT AMX XI VAI 75).
ON72 14ND X21] lacking in manuscript M of b. Meg., y. Meg., Mek., Abot
de-R. Nat., Soph. and Yal. Shim.
5. OnDv OW 77981 TIN Tan (Gen 11:7)
onow ow 772x1] lacking in y. Meg. and Abot de-R. Nat.
6. WaNPa Mw pnxm (Gen 18:12)
y. Meg., Mek., and Soph. add: 70x.
7. DIAN PY ONE) Ww 77 ODNI °D (Gen 49:6)
wv] manuscripts of Mek. and Soph.: wx
8. DUN 71D NW PY ODT 119: NN NWN NX Tw 4 (Exod 4:20)
1712 Np] missing in Abot de-R. Nat.
O71 712 xwi] manuscript M of b. Meg., Mek., Midr. Hagadol Exod.
Abot de-R. Nat. and Soph.: 078 (*)xwa.
xwi] y. Meg. and Yal. Shim.: °xw1.
9. YIDNI Aw Ow?wW MSN INWI ON Ww? WR ON IW? 73D Iwi)
mw MN” (Exod 12:40)
o-7xn3] Soph.: 28) pruRXa
mya awn] y. Meg.: nyzqN7 292); Midr. Hagadol Exod: mxnKxn 7AKNwI);
Mek.: ]W(1)) YAX2) 1V1D PIXI1, so also Tan. in inverted order; Soph.: 1915
Y1N21; mss of Soph.: O39 PARI) 7V3ID PAN2.
10. 2xqw? 722 -DION! 17w7)1 (Exod 24:5)
*v1UNT] Midr. Hagadol Exod: *y1vy1; Yal. Shim.: *v101. The entire
passage is lacking in y. Meg., Abot de-R. Nat. and Soph.
11.0 ndw x2 Ox w 212 ~wNt DN? (Exod 24:11)
>vIUNT] Midr. Hagadol Exod: *vivy1; Yal. Shim.: >v101. The entire
passage is lacking in y. Meg., Mek., Abot de-R. Nat. and Soph.
12. -NXwW) OF INN TAN NX? (Num 16:15)
ton] Mek.: an. Tan. lacks the entire passage.
13. omy 292 PRA? OME ADRK ‘nH ?9n AWE (Deut 4:19)
WRxI?] Abot de-R. Nat. adds: 073. y. Meg., Abot de-R. Nat., Soph.
and Tan. add: a-nwn (3) nnn.
14. o73Y9 -n8 XR? OWE ONN ODN T7179) (Deut 17:3)
o-anKx-727)] lacking in y. Meg., Mek., Midr. Hagadol Exod and Tan.;
ms M of b. Meg. and Yal. Shim. omit o-7nK OAK.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 5
o7ay9 7] Soph.: o72192 “ne RX? WR OWA RIX 799 NK 179 IR ww);
Mek.: 072972 NN? "MPI NX? WR I? 14ND.
oway?] y. Meg., Mek., and Midr. Hagadol Exod: (1729?) o7Aay? niniK?.
15. NI0N7 NN IND NP aPIAA Nw 9? 14ND (Lev 11:6 (5); Deut 14:7;
the continuation of the passage is quoted below).
nainwni— ans] y. Meg.: on mye AX nN nx; Mek., Soph. and
Yal. Shim.: 07177 n-wyx nx(1). In y. Meg. and Soph. this passage comes
after passage 9; in Abot de-R. Nat. it comes after passage 12.
3. The circumstances under which the alterations were inserted in the
EXX
The circumstances under which the alterations were inserted in the
LXX are described in the introduction to the list, whether in brief or in
detail, and the name King Ptolemy, ‘for’ whom the translators ‘wrote’
their translation, is mentioned in all the descriptions.
The short descriptions speak only of ‘writing,’ as in Mek. (‘and this is
one of the things they wrote for King Ptolemy. Similarly they wrote
him ...’) or of an ‘alteration’ as in y. Meg.: ‘thirteen details were
changed by the sages for King Ptolemy; they wrote for him ...’.
The longer descriptions relate the story of the writing of the LXX
known also from other sources, both Hebrew and Greek,? although the
differences in outlook and emphasis between the rabbinic account and
the other sources are considerable—see Aptowitzer, “Berichte” 3 (1910)
4 ff. B. Meg. relates the following account: ‘It has been taught, the story
goes that King Ptolemy assembled seventy-two elders and lodged them
in seventy-two rooms without disclosing to them the reason for
assembling them, and he went into each one individually and ordered
them “write me the Torah of your Teacher Moses.” The Holy One,
blessed be He, put wisdom in the heart of each one so that they agreed
with one accord and wrote for him ...’ (at this place follows the list of
alterations).
This account describes the circumstances under which the Greek
translation of the Torah was prepared, and if not all the details of this
story are mentioned in every single source, it is often alluded to in such
phrases as ‘they wrote for Ptolemy.’ Furthermore, Midr. Hagadol Exod
4:20 says explicitly: ‘this is one of the eighteen details which our
Rabbis changed in the Torah in Greek.’ Significantly, in Soph. 1:7 this
> See P. Wendland, Aristeae ad Philocratem Epistula cum ceteris de origine versionis LXX
interpretum testimoniis (Lipsiae 1900); H.St.J. Thackeray, The Letter of Aristeas, Translated with
an Appendix of Ancient Evidence on the Origin of the Septuagint (London 1918).
6 CHAPTER ONE
story is mentioned together with another one which speaks explicitly
about the circumstances in which the LXX was produced (‘Thus goes the
story about five elders who wrote the Torah for King Ptolemy in Greek
etc.’; after it comes the story under consideration here beginning with
the words ‘Another story about King Ptolemy ...’).
4, Writing or alteration?
A few traditions speak of the ‘writing’ of the above-listed passages,
while others speak of the ‘change’ from the Torah (see above). It would
seem that even if it is not stated explicitly that the sages/elders/our
Rabbis inserted alterations, such a claim is inherent in the very
formulation of the list. First, all the passages mentioned in the list
differ from MT. Second, for two passages the content of what the
translators wrote is explicitly stated instead of other details: 4 ‘male
and female he created him’ and they did not write ‘he created them’
(Gen 5:2; the final three words are lacking in many sources); 15 and they
wrote for him 07717 nv"yx and they did not write n217x (Lev 11:6 [5], Deut
14:7; the various traditions differ, but all of them refer to both
expressions in one form or another).
Thus, the story preserved in rabbinic literature records the
alterations from the Torah inserted by the translators. It was only
natural that people should soon recognize the existence of differences
between the Hebrew and Greek Pentateuch. The latter, too, was
‘Jewish’ at its source, even though the Jews distanced themselves from
it at a later date. Furthermore, it was also natural that every
difference between the Hebrew Torah—being in the language in which
the words were originally written—and the Greek Pentateuch should
be thought of as an alteration in the Greek. The real background of the
aforementioned differences between the Hebrew and the Greek
Pentateuch is dealt with below. Apparently, some of these differences
do indeed stem from alteration, but others, probably the majority, stem
from Hebrew variants, from translation technique and from an incorrect
under-standing of certain translation equivalents in the LXX. All the
same, the differences mentioned in the list as ‘alterations’ are
described as such here, because this is how rabbinic tradition
understood them. Christian tradition also took similar differences
between the ‘Jewish’ and ‘Greek’ (from their viewpoint: Christian)
Bible to be alterations, but in the opposite direction: a few Church
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 7
Fathers claimed the LXX reflects the true form of God’s words, and that
it was the Jews who had falsified them in their Bible.®
5. The original language of the passages mentioned in the list
The list contains a number of altered passages, inserted by the
translators and differing from the Torah—thus according to rabbinic
tradition—and it can indeed be verified that all the passages differ
from MT. Therefore the passages listed in Hebrew refer to the Greek
translation of the Torah, which is quoted in the list in Hebrew
retroversion. Interestingly enough, a few researchers hold to the
opinion, for reasons which will be treated later, that these are not
citations from a Greek translation at all, but rather alterations on the
Hebrew level.’ This opinion does not appear likely, however, in view
of the fact that the introduction to the list explicitly refers to a Greek
translation. In addition to this, from some details in the list it also
emerges that the citations come from a Greek translation:
1. Five of the passages are identical to passages in the LXX (3, 8, 10,
11, 12, 15), with another one (9) being close to it.
2. The supposition that the list goes back to Greek words that were
translated here into Hebrew is well substantiated by passage 15. There
it is said that the translators wrote 07717 nvyx (young-footed) ‘and
they did not write na1nKx (hare) since Ptolemy’s wife’s name was ‘hare,’
that he might not say ‘the Jews have mocked me by putting my wife’s
name in the Torah’ (b. Meg.). In fact, the people did not nickname
Ptolemy’s wife (actually his mother) n210x, but instead used a Greek
equivalent (Aaywdés). Therefore, if nainx refers to Aaywés, the phrase
o-o17 Nyx points to nothing else than a Greek word of equivalent value.
Indeed, it is possible to identify the Greek word behind 07737 nvvz: the
Greek equivalent for n217" in the LXX of Lev 11:6 (5) and Deut 14:7 is
Sao(toSa, whose meaning is ‘hairy-footed’ (0°71 nvyw). Undeniably
this is the phrase 07717 nvvyx in the words of the sages, presented thus
by a phonetic interchange of x/ w.8 Furthermore, the equation of nvyw
6 See, for example, Justin Martyr, Dialogue with Trypho, 3.1; P. Benoit, “L’Inspiration des
LXX d’aprés les Péres,” Mélanges H. de Lubac, I (Paris 1963) 169-187.
7 Frankel, Vorstudien, 31; Friedman, 23 ff.; Talmon, “Scrolls,” 26. Aptowitzer, “Berichte” 2
(1909) 7 ff., rejects this view.
Cf., e.g. Num 16:30 7nx51 as against the reading of the SP anwo,; 2 Sam 8:3 2°wn as
against 2°x79 in 1 Chr 18:3. See also 1pn°w in the baraita itself and cf. for this issue A.
Bendavid, Biblical Hebrew and Mishnaic Hebrew 2 (Tel-Aviv 1971) 441 (Hebrew). The
identification of a°717 nv-yx with ov717 ny" was first made by G. Tychsen, Tentamen de
variis codicum hebraicorum ... generibus (Rostock 1772) 52. Tychsen also discusses the rabbinic
tradition.
8 CHAPTER ONE
0-717 with SactTo8a appears reasonable in light of what is known
about the use of compound words in the LXX and about the translation of
these words in Hebrew and Aramaic: many pairs of two or more Hebrew
words are translated in the LXX by compound words of the type fw? 735
- Bpadt'yhwoaog (Exod 4:10)—see Tov, “Compound Words.”* Alternati-
vely, compound Greek words were many times translated by a phrase of
two Hebrew or Aramaic words, as can be recognized for instance in the
Syro-Hexapla.? Moreover, the translation of Sac¥to8a in rabbinic
literature needs to be seen in the light of the LXX vocabulary in which
Tove generally reflects 717 and Saovc reflects yw as in Gen 27:11 (cf.
also Gen 25:25; 2 Kgs 1:8).
3. The assumption that the passages mentioned in the list reflect
Greek and not Hebrew words emerges also from passage 12: 79n
mentioned there reflects émv6Uprpa in the LXX (MT: 7inn). Within the
LXX the root 1m is generally translated by émL8up-, and so 7nN is
translated in Isa 32:12 by ém.6vpnpa. Therefore the reconstructed process
ton (the conjectural origin of the LXX) = émLO0pnpa = Tan (= the rabbinic
list) points to a translation process.
4. Ostensibly, the change of word order in passage I (R12 MwxNrI
a°n?x) and the expressions 8 07x 712 (*)xw1l = bTofvyta and 7 012 =
oiTeuTog (see below) can only be understood by the assumption that
these are translations from Greek.
6. The list of alterations and the original text of the LXX
In the past, when scholars observed that the list contains passages
which agree with the LXX, they shirked from applying this
description to the entire list, since the majority of its details go against
the transmitted text of the LXX. A comparison of the passages with the
LXX shows that nine passages in the list differ from the LXX, while
five agree with it (3, 8, 10 , 12, 15), with one passage being close (9).
If the preceding analysis is correct, it is difficult to avoid the
unusual assumption that the nine passages which do not agree with the
transmitted text of the LXX reflect another textual form of that
translation. This other text of the LXX evidently contained the
original text of the translation which differs from the transmitted form
in all the other manuscripts. This assumption is strengthened by what
is known about the textual development of the translation during the
first centuries of its existence. This question is now briefly considered.!9
9 E.g. Exod 4:10: pw? 7259 75 123 - layvdéguvocg kal BpaStywooos - 821 171 827? Win
ae passim in the LXX: Sry WwP - axdnpotpdyndros - X27 NWP.
10 See further, Tov, TCU, 10-15.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 9
It is reasonable to hypothesize with P.A. de Lagarde, Proverbien, 1+,
that the manuscripts of most, if not all, Septuagintal books, reflect in
one form or another the first formulation of the LXX, which we may
denote for the purpose of discussion as ‘the original translation.’ This
original translation was not preserved in its pure form for an extended
period because from the beginning of its dissemination in different
scrolls, the textual transmission split off into several secondary
traditions. In the pre-Christian period and the first century CE various
types of corrections were then entered into individual scrolls of every
one of the Septuagintal books. As a result of these corrections, as far as
one can tell, there were no two identical or nearly identical scrolls in
existence for any book of the LXX.!! In contrast to this situation, by the
second and third century CE, a recognizable unity had come about in the
textual tradition of the LXX which later disappeared under the
influence of the revisions of Origen and Lucian.
For the present discussion it is important to know which types of
alterations were inserted in the textual witnesses of the LXX. The
evidence shows that many alterations were inserted in early witnesses
which brought the LXX into conformity with the Hebrew Bible. Some
revisions were inserted in the forerunners of the translation units now
found in the canon of the LXX,!2 while others are reflected in
individual manuscripts, such as manuscripts AFM in Exodus-Deuter-
onomy.!% Furthermore, even if in a certain detail all manuscripts of the
LXX agree with MT, there is no certainty that the original translator
indeed produced this rendering, because the original rendering may
have been corrected in accordance with MT. This assumption received
support from 4QLXXLev,!4 which sometimes reflects a text which is
probably original, while the transmitted text of the LXX was probably
corrected toward the standard vocabulary of the LXX and/or MT.!9
11 This point was emphasized by E.J. Bickerman, “Some Notes on the Transmission of
the Septuagint,” A. Marx Jubilee Volume (New York 1950) 149-178.
This situation is recognizable, for example, in the “LXX’ of the following books: parts of
Samuel and Kings, Daniel, Ruth, Ecclesiastes, Canticles.
13 Dw. Gooding, Recensions of the Septuagint Pentateuch (Tyndale Lecture 1954; London
1955),
14 See the discussion of P.W. Skehan, ’The Qumran MSS. and Textual Criticism,” VTSup
4 (1957) 155-160 and of E. Ulrich in DJD IX, 161 ff.
The alternative view, according to which the scroll reflects an early revision towards
a freer rendering of MT, is not borne out by the evidence.
10 CHAPTER ONE
In view of this situation, it is suggested here that the passages
mentioned in the list of alterations reflect the original text of the LXX,
while the archetype of all the known manuscripts was corrected.!¢
As for the frequency of the presumed corrections of the original text
of the LXX, the assumption that two-thirds of the passages in the list
were emended in the archetype of Septuagint manuscripts is not
illustrative of the frequency of such changes, which must have been less
frequent.!”
We now turn to the ten passages differing from the transmitted text
of the LXX; their original form will be reconstructed on the basis of the
rabbinic tradition. The discussion includes passage 9, which agrees
with the LXX to a limited extent.
The tentative retroversions from the Hebrew of the list to the Greek
of the LXX are based primarily on the vocabulary of Hebrew-Greek
equivalents which served the translators. These reconstructions
encounter the same methodological difficulties as do retroversions in
the reverse direction. The degree of reliability of the reconstruction
depends on the degree of exactness in the translation. It should
therefore be emphasized that the Hebrew translation in the list of
Greek passages appears to be exact. This exactitude is recognizable in
the literal translation of the two elements of SaovT0Sa (15) by nvyw
o-2)17 = 07717 ny (see below) and in the translation from the Greek
(possibly: tot katpevetv attoic) reflected in 0729? (14)—such a reading
is indeed reflected in a Hebrew source (Siphre Deut 19:19). It seems
that only in one biblical passage is a Greek word presented by a free
translation: O7X 712 (7)Xwi = UTocuyra (8). If this description proves
correct and the Hebrew translation in the list is indeed literal, our
reconstruction stands on a firm basis. In fact, the very nature of the list
demands that the translation incorporated in it be exact, since the list
purports to faithfully represent the differences between the Torah and
the LXX.
We now present a tentative reconstruction of the original text of
those passages in the list which differ from the transmitted text of the
LXX, accompanied by remarks on the retroversions. The transmitted
text of the LXX is recorded first, followed by the text of the LXX
reconstructed from the rabbinic tradition. These passages have now
been analyzed in detail by G. Veltri, Eine Tora ftir den Konig Talmai—
16 Absolute originality cannot be proven. In our view, the passages in the list reflect a
text which is more original than the ones in the known manuscripts of the LXX.
7 We are faced with a list of differences or changes, which are not characteristic of the
general condition of the text.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 11
Untersuchungen zum Ulbersetzungsverstandnis in der jiidisch-hellenis-
tischen und rabbinischen Literatur (TSAJ 41; Tubingen 1994). The focus
of this detailed study differs from our study and in a way the two
studies complement one another. See also Tov, “Review of Veltri.”*
1 Genl:1LxXxX ~— év dpyxij étrolnaev 6 bed6¢
=MT ODN NID MWR ID
LXX-reconstr. 6 9€d¢ étolnoev év dpxTj
=rabb. list =m-wx ia x12 O70RX
2. Gen1:26LXX totjowpev dvOpwrov Kat elkdva hWEeTEpav Kal
Kad dpLoiwoty
=MI WNT W978] OF WI
LXX-reconstr. tTo.fjow dvéputrov kat elkdva kai ka® dpolworv
= rabb. list n72) 07¥2 O7X NWR
One of the two differences between the LXX (= MT) and the
reconstructed LXX (= rabb. list) concerns the person of the verb (see
below). The reconstruction does not relate to prepositions in the list:
3...2 (MT 3...1; LXX apparently 3...2), because this type of difference
cannot be reconstructed for the LXX. The other difference between MT (=
LXX) and the retroverted LXX is based on a reliable tradition.
4. Gen5:2LXX dpoev xal 6jAvu étrolnoev auTouc
= MT ONT APN D1
LXxX-reconstr. dpoev kat OrjAu étrolnoev autév
=rabb. list IW 1373) DT
The reconstruction is based on the text of b. Meg. See also n. 29.
5. Gen11:7 LXX deiite kal KataBdvtec ovyxéwpev éExet avTay THY
YAuUCoav
=MPF onow ow a22n WIT AIA
LXxX-reconstr. Seite kal kataBdc ovyyxéw...
=rabb. list onow ow 77aN7 TITAN AAA
6. Gen18:12 LXX éyédacev & Zappa év EauTy
=MT M42 Ww PpnXm
LXX-reconstr.(?)éyéAaoev 6€ Zappa év/tpdq/étml Tote /TouG
éEyyLoTa auTyc
=rabb. list M2np2 Ww prxm
The difference between the reading of MT (= LXX) and that of the list
(7°217P2) may be explained as follows:
1. If 772972 in the list refers to people standing near Sarah (see the
early commentators on the rabbinic list) or to her relatives, the
12 CHAPTER ONE
meaning of the passage is that Sarah laughs at these people. In this
case the original text of the LXX may be reconstructed as above.
2. Most modern interpreters hold that the difference between the
passage quoted im the list and MT does not bear on the quoted words, but
rather on the continuation of the biblical passage. Indeed, in the
continuation of the sentence, the LXX (ottw év jot yé-yovev Ewe Tot viv)
differs in three details from MT (a17y 77 ANA 7N?2 7TNRX): "INN is not
represented in the translation. Instead of °n?1 the translator read *n7a,
and instead of 737y he read AI3y (= 710 Ty).
3. Possibly the two words differ solely in their pattern (027?12/
711772), their meanings being identical—cf. the transcription of 1773 by
BekopB in the second column of the Hexapla in Ps 36(35):2 and notice
similar phonetic shifts in mishnaic Hebrew.!8 Also the MT of Isaiah
and 1QIsa@ differ in many instances as to noun patterns!? and such
differences are also to be assumed at the base of the relationship
between MT and the transcriptions in the second column of the
Hexapla.*9 But even if 7°21772 reflects a different pattern of the word in
MT, the original translation should probably be understood as “people
standing nearby’ or ‘relatives.’
7. Gen49:6 LXX étt év Ta Ou avTav atéxTeLvav dvOputrous Kal
év TH émLOupla abtuv Evevpokdtmoav Tatipov
(Tavpovc manuscripts 458 340...)
= MT Ww PY OID) WX WIT DDR 7D
LXX-reconstr. ... €vE uxpoKO6TOQV OLTEUTOD ...
rabb. list OIAX py (0872) Ww 77 ODN 7D)
The point of departure of the reconstruction is 013x*! (01x = fattened =
o.tTeutToc) which appears in all sources of the list (in most of the lists
“iw comes in the first hemistich, while in some of them w°x appears as
in MT [see n. 30]). An examination of the translation equivalents of the
LXX shows that 0128 in the list may reflect outeuto¢ which in the LXX
also translates Tw (that is, 012% in the list = otte€utdéc¢ in the recon-
structed LXX = 71v in the Bible). This assumption is based on the
following equivalents: Judg 6:25 wi 19 - Tov pdcyxov TOV OLTEUTSV
18 Soe G. Mercati, Psaltertt Hexaplt Reliquiae (Roma 1958).
19 See Kutscher, Language, 396-398.
20 See E. Bronno, Studien iiber Hebriische Morephologie und Vocalismus, auf Grundlage der
Mercatischen Fragmente der zweiten Kolumne der Hexapla des Origenes (Leipzig 1943); Z. Ben-
Hayyim, Studies in the Traditions of the Hebrew Language (Madrid/Barcelona 1954); A.
Sperber, A Historical Grammar of Biblical Hebrew (Leiden 1966).
1 The vocalization of the Adler manuscript (0324), like the orthography of manuscript
Columbia X 893 — T 141 (012°x), is apparently secondary.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 13
according to manuscript A (cf. the text of B: Tov pdaxov Tov Tavpov); 1
Kgs 5:3 O°012N... IRS - TpdBata ... oLTEeVTG; Prov 15:17 ... DIN Nwn - ...
ite p Bowv o.teuTév. 24
9. Exod 12:40 LXX fh 6 katolknote Tov ulov Iopana fv KaTwWKnOAaV
év yj Alyvmtw kal év yj Xavaav é™m tTetpakdéova
TpLadKOVTAa
MT YIN) Tv aww ox. aw? Wr bRaw 3a awin)
AW NINA
reconstr. (a) é€v (yj) AlydtrTw Kal (Ev) tate katadoltaLe
XWpaLc (TuV xupwv)
=rabb. list niy7X8 TXWI) OTA
reconstr.(b) = év (yf) AlytmTw kal év tdotv Tate yaparc
= list in y. Meg. miy0KNF 732) O32
reconstr. (c) év (yj) Alyutrtw kal év tate xwpate taic ETéEpate
reconstr.(d) év yj AlyvTTw kal év Xavaav kal év yR Tecew
Reconstructions a—c are founded on the assumption that the translation
in the list is exact. Whether the majority reading 1xwW2) is more original
than nyxiNx7 7201 in y. Meg. cannot be determined. Both readings could
have stemmed from étépat¢ in reconstruction c. It should be observed
that nixqX xv does not occur in the Bible and the absence of the article
in NX points to mishnaic Hebrew (against this, manuscript Columbia
X 893 — T 141 of b. Meg. reads mx7N7 73 ANW)).
Reconstruction d is based on the assumption that the Hebrew
translation in the list is free. The Greek text may then be reconstructed
according to the LXX of the passage, according to the reading in Mek.
(wi POND) 791D pqxd) o73N2; thus also Tan. with inverted word order),
and according to the SP ad loc. ("78 PANI) J¥1D PAN).
11. Exod 24:11 LXX kal Tuiv ETLAEKTWV Tot Iopand ov bleduvacev
ovsé ele
MT TD onde@ xd Ox we 71D MRR OR)
LXX-reconstr. ... Tov veavlokwy / Tov EhATTOULEVWD? ...
=rabb. list (?x7w7 713) *WION? ...
Passage 10 is treated above together with Exod 24:5 similar to the
present passage. Since DivNT-DIDYT is attested in Hebrew and Aramaic as
‘small,’ it probably goes back to veavfoko. in the LXX of v. 5. In
accordance with passage 11, this word probably appeared in the LXX of
our passage, but here the problems are more involved than in v. 5.
22 attributed to Aguila and Theodotion; a similar translation is handed down as e€’.
14 CHAPTER ONE
From a linguistic viewpoint it is possible that the Greek translator
would translate *?°xx with a Greek word which would be retroverted in
the list as bioNT. Certainly the root xx, which is related to the
preposition 7?Xx, appears both in the sense of ‘to lack’ (Sir 42:21) and in
the sense of ‘to set aside’ (Gen 27:36; Num 11:17). From this it can be
conjectured that the original equivalent of *?-xx was édXatToupévwy, cf.
Sir 42:21 23x12 x21 [F/O N2] - o} TW TpCGETEHH OTE HAATTWON (NO NX> is
retroverted according to the LXX there, and support is now forthcoming
for this retroversion from MasSir, which reads 0x1 8? [9x1 ¥87)]).
Consequently the following process is reconstructed: *?>xx in the Bible =
EXATTOUWEVWY in the reconstructed LXX = *vivnt in the list. But it is also
possible that veavfoKxot appeared here, as in v. 5, if the translator
identified 2x7w> 719 *D>xN (v. 11) with Dx iw? 712 “41 (v. 5).
The original equivalent of *?°¥x cannot be reconstructed easily
because the construction of the verse differs entirely in its Greek
translation. Actually, *?°xx is represented twice: (1) émAéKtTwv (chosen
ones) reflecting the accepted interpretation of *?°xx, and (2) it is also
concealed behind 8teguivnoev: the Greek translation of "12 "73x ON)
17 nw RX? INIW? should be understood as ‘and from the chosen of Israel
not one was lacking.’” Because the root 0¥X means ‘to lack,’ Steduivnoev
somehow reflects ?xx (rather than 17 n>w x> or a Hebrew variant).
Alternatively, the translation may express tendentious exegesis, as
elsewhere in the immediate context.2* To reconstruct an individual
detail in the verse is all the more abstruse.
13. Deut 4:19 LXX 4 dtéve.pev ktptog 6 bed¢ cou atta Tdotv Tot¢
E8VEOLV TOLG UTOKATW TOU OUpavot
MT Oenwn 2D nnn ones 359 onx WAN ‘0 Pp9n AWK
LXxX-reconstr. ... auTd + Tot dutlCeaber + TdoLy €Oveoiv
rabb. list + PNT,
The equivalent xn - dwttCecGar also occurs in Num 8:2; Ps 13(12): 3.
14. Deut 17:3 LXX kal dted@dvtEec AaTpevowoty Beoic ETEpoic .. A
ov tTpooétaéa (-Ee(v) BG...; = cot FMV...)
MT “NY NP IWN... DANN OTR Tay 72%
LXX-reconstr. ... EtEpo.g + Tot AaTpeveELv abToie ...
rabb. list +o 72?
23 See the similar use of the verb Stadwvéwin LXX Num 31:49: wx nn 77D) X21 - Kal ov
Btateduvynkev an’ abray obSé elc; Josh 23:14: InX 737 491 XY - ob Bredubvnoev EE adTuv; see
also 1 Sam 30:19; 1 Kgs 8:56.
4 See especially v. 10: 28QWw* “72X NX NT - Kal El8ov Tov Térov of eLaTHKet (Exet) 6 BEd
Tov Iopand and v. 11: oF>XT RX WN - kal Ghdonoav év TH ToT TOD BEoU, for which cf.
Ch.T. Fritsch, The Anti-anthropomorphisms of the Greek Pentateuch (Princeton 1943) 45.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 15
The equivalent 712y - AatpeveLv occurs among other places in Deut 28:14;
Judg 2:19. |
See also below on the earlier formulation of the LXX to passage 15.
7. The background of the differences between MT and the LXX
The lists in rabbinic literature speak of alterations inserted in the
translation, but this notion may now be abandoned.*° The differences
between the Torah and the LXX derive from: (a) translations deviating
from MT based on Hebrew variants; (b) translations deviating from MT
arising either from Hebrew variants or from exegesis; (c) exegetical
translations; (d) Greek equivalents which were misinterpreted by the
rabbinic tradition as differences between the LXX and the Torah.
The contents of lists of this type are largely a matter of chance, as is
also the case with the list of the ‘“emendations of the scribes’ (see n. 1).
This list does not purport to represent the most conspicuous alterations
and indeed anyone will easily find much more far-reaching differences
between the LXX and MT, as for instance in the order of chapters and
subject matter at the end of Exodus. What the passages in the list have
in common is that they pertain to some central issues. These differences
could easily be reinterpreted as alterations (like the ‘emendations of
the scribes’).
a. Translations deviating from MT based on Hebrew variants
The following passages most likely reflect Hebrew variants:
12. Num16:15 MT cnxw) onn InX tn KX?
LXX-reconstr. °nxXw32 070 78 nn KX? = list
Possibly the reading 79m was created when a copyist or translator
replaced 1mn with 7v9n.2® Cf. Gen 49:14 O71 Won - TO Kaddv ETTEBULQGEV =
o71 1())on (cf. similar linguistic exegesis in the LXX of Ps 119 (118):20
Ton) - éteT6Onoev)27 and Isa 27:2 ann 07D - duTedAQw Kaddg ETLOUpNLA ...
25 Some analyses of the differences between the rabbinic tradition and MT, like the
traditional interpretations of the rabbinic tradition, presuppose that all the variants
discussed reflect changes made by the translators. See, e.g., Geiger, Urschrift, 282-287; A.
Kahana, o-nN"nA O1D07 2 (Tel Aviv 1960) 16-17; M.H. Segal xan xan 4 (Jerusalem 1960)
928-930.
26 According to rabbinic tradition, both here and in passage 8 (Exod 4:20) the translation
does not list an in relation to Moses, but this situation does not support the assumption
that_any bias is evident in the translation.
It is difficult to assume that this interchange was tendentious; it is more probable that
it derived from difficulties in identifying the rare word 07) (cf. also LXX of 2 Kgs 9:13).
16 CHAPTER ONE
MT apparently reflects the original intention of the text: Moses
emphasized that he has not even taken for himself a small thing such
as an ass. His words are similar to those of Samuel just before his
death: *nn?? * vm (1 Sam 12:3 MT LXX).
13. Deut4:19MT ony 2D? TDN 17 PON WR
LXX-reconstr. a “ pope % “ % ” =rabb. list
The added 1°x7? is also reflected in Midr. Hagadol to this verse (see
also Rashi). This word evidently testifies to a tendentious early
variant: one is not permitted to worship stars: the host of heaven are
given only ‘to shine.’
14. Deut 17:3 MT IE NOW ... DANN PR TV 77}
LXX-reconstr. ovay>+ “ ” ” u “ma
= rabb. list
The additional 07129? may reflect an early variant echoed in Siphre 148
on Deut 19:19. Cf. also Deut 28:14 o72y? ON DTPX MINK N37?.
The addition is occasioned by linguistic considerations because Wx
n-1¥ NX? has a certain syntactical oddness about it and begs, as it were, to
be completed. The additional word is taken from the context (7297)
ovay> + °n 7X XN? WRX O7NNX O°72N); early commentators added the same
word.
b. Translations deviating from MT arising either from Hebrew variants
or from exegesis
In certain categories of translation technique it is difficult to determine
whether a specific difference between MT and an ancient version attests
to a Hebrew variant or reflects the exegesis of the translators (see TCU,
154-162). A few of the passages in the list belong to this group (note
that the list did not intend to reconstruct the Hebrew original from
which the LXX was made, but rather to communicate a ‘contemporary’
Hebrew translation of some points of interest in the LXX).
2. Genl:26 MT nin 199782 O78 Aw)
LXX-reconstr. Tottjow dv@pwiov kat elkéva kal ka& dpolworv
(rabb. list m7) 0°83 ON TwYR)
5. Genil:7MT Avan ata nan
LXX-reconstr. Sette kal kataBdac ovyxéw (list 59281 79708 17)
In these two instances the translation avoids adopting the plural as in
MT. The translator either inserted this alteration on his own
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 17
initiative, or he found such a Hebrew text in front of him. It is pertinent
to remark here that in b. San. 38b, Gen 1:26, 27 and 11:5, 7 are cited
together as examples of the use of plural forms in reference to God.
Besides the difference in the person of the verb, an additional
difference is noticeable in passage 2: according to the list, the
possessive pronouns in MT are not transmitted in the LXX (n721 07x21 as
against 11N%7D 119982 in MT). This lack is possibly based on an early
Hebrew variant, for also in rabbinic literature 07% and nin7 are
sometimes described in an abstract way, without possessive pronouns or
article (note, for example, Abot 3:21 0982 °3 Woxiw OF82 XIWIW OFX an
OINT NX AWY arn>N). It is also possible that the translator did not
represent the pronouns in order to avoid an anthropomorphic
description, as elsewhere in the LXX.78
According to the list in Midr. Hagadol Exod 4:20, the pronouns are
lacking in v. 27 of the LXX as well: n973) 0933 O7N7 NX OR NIA. If the
tradition concerning the Greek translations of v. 26 is correct, it stands
to reason that v. 27 was formulated in the same way.
3. Gen2:2 MT 9 cyrawa or. o7bxX 25)
LXX kal ovvetédeoev 6 Bede Ev TH hepa TH EKkT™m
(list *wwn ara 257)
Ancient as well as modern interpreters were aware of the exegetical
difficulty raised by MT, which implies that God worked on the
seventh day. This difficulty is eliminated in the LXX. The Greek
translation may be based on a Hebrew variant (‘ww7) also found in SP,
but it is equally feasible that the translator changed the content of the
verse.
4, Gen 5:2 MT ON] 13/3) 733
LxXX-reconstr. dpoev kal 6rjAv étrolnoev avtév
(list: INT2_713PN 71)
It is hard to know whether the reconstructed translation éto(noev
abtov reflects a variant 1X72 or results from exegesis. Either way, at the
base of the translation one should probably posit an understanding
which interprets this verse as referring to an androgynous creature as in
Gen Rab. 8:10. However, possibly the variant or the tendency lying at
the base of the translation did not stem from contextual exegesis, but
from syntactical exegesis since the previous verse speaks of the man in
28 See Fritsch, The Anti-anthropomorphisms, 11, n. 6 with reference to omissions of the
possessive pronoun in relation to God (LXX Exod 15:7; 23:27; Deut 32:10). In other cases an
element is added in the translation between two words in order to tone down an
anthropomorphic description.
18 CHAPTER ONE
the singular, while v. 2 speaks in the plural (08727 O72 O7X Onw NX NWP)
onix 772"). The passage under consideration comes between these two
passages and hence it causes no surprise that in some source a variant or
an explanation of 1X72 would develop for MT ox 72. The variant 17273)/
172371) included in many sources of the list apparently reflects a
secondary stage in the development of that list.29
ee Exod 12:40 MT vary) aw o-w>w) oyna (1a? WwE DxIw? 722 3w1nN))
(lw Moxa
LXX-reconstr. see the four possible reconstructions on p. 13.
It is unclear whether the difference between MT and the LXX is to be
attributed to an exegetical alteration or to a translation of a Hebrew
variant like that appearing in SP (Q°-73m pox) 7V12 pra) or in Mek.
(wi PANT) 7VID pIN3A O'7%N3; thus also in Tan. with inverted order).
c. Exegetical translations
1. Gen1:1 MT OT? NAD MwWNAD
LXX-reconstr. 6 Oed¢ évolnoev év dpyy (list: n-wWRID NID O72N)
The LXX translators often inverted the order of elements whether from
syntactical or exegetical considerations. In this instance the inversion
can be ascribed to the translator’s motivation to begin the translation
with 6 Gedc.
d. Greek equivalents which were misinterpreted by rabbinic tradition
as differences between the LXX and the Hebrew text
In the following four examples the translators chose regular equiv-
alents which in rabbinic tradition were misinterpreted as reflecting
differences between the LXX and the Hebrew text.
7. Gen 49:6 MT TW WPY OFNI3) WX WI DDR3 7D
LXxX-reconstr. ... €veupoxétmmoay ottevTév = list DIX Py ...
oLTeuTdg (= 072X) in the LXX was apparently interpreted as an alter-
ation of the parallel word in the Hebrew (71w) since it was generally
understood as 711 = wall (T°, S, Aquila, Symmachus, V; see also Gen
29 4 reading 12/71/121?1 was probably created during the textual transmission of the list
(interchange of 7/1(’) perhaps when it became unclear what the exact difference between
LXX and MT was (a similar confusion is reflected in the addition of ox73 12ND X71 in the
list, a formula recurring only concerning passage 15). It appears that 12?1/ 12173 refers to
the female orifices of the primeval man who was thus androgynous (see Gen. Rabb. 8:10).
Worthy of note is the fact that 2171/2?) does not appear in any similar meaning in the
Bible, and in rabbinic literature it is used only in connection with the male sexual organ.
THE RABBINIC TRADITION CONCERNING THE ‘ALTERATIONS’ 19
Rab. ad loc.). Consequently, according to rabbinic tradition, the LXX
reflects an alteration, but actually the translation is based on a reading
tradition which is also reflected in MT (riw).32
8. Exod 4:20MT onn
LXX Ta UTocUyta (list OFX 712 (7)xw11)
If 078 "322 (*)Xwi is a faithful translation of the original Greek trans-
lation, it may be reconstructed as *dvOpwtoddpoc or *PopdvOpatoc,
though neither of these words is attested in Greek. Apparently in this
instance the Hebrew translation in the list is imprecise, meant to
emphasize the interpretative dimension of the Greek word. Etymolo-
gically the meaning of Uto{uy.ov is a ‘yoke (¢vydv)-bearing animal,’
and as such designates various animals. On the other hand, in Egypt its
meaning is restricted to ‘ass,’ or at least this meaning was prevalent
there.?! For this reason 710m is often translated in the LXX by ttoftytov
(the equivalent Jin - dvoc is more frequent in the LXX, with the
exception of Exodus in which the passage under consideration
appears).3* We may infer that this use of Uto{ty.ov was unknown in
Palestine, so that the sages were puzzled about the use of O78 712 (*)Xwi]
(this is how they understood tro{tytov) and not 6voc, as usual.
10. Exod 24:5MT (Ox tw? 733) “WI
LXX tTovc veavloKoug = list *DIUNT
The equivalent 1v1 - veavlokog/veavlac occurs frequently in the LXX
(matSdptov is more frequent). Therefore the use of veavloxog need not
have raised any difficulty on the linguistic level, but on the exegetical
level it was apparently considered unusual, because these O°191 were
usually taken to mean ‘select men’ (see, for example, T° --192 and Zeb.
115b) and not ‘little ones.’ Against this background the sages may have
30 The main differences between LXX and MT pertain to 0128 and 71. But in the
parallel (first) colon tw appears in several lists instead of MT’s w°x (= dvOpaitious in LXX).
Apparently, 11@ is secondary here and was entered under the influence of 11 at the end
of the verse (where it was replaced by 0122). It is unlikely that the original list intended to
ascribe this variant to the LXX. See also the preceding note.
31 See the lexicons and particularly MM with examples from papyri of the 3rd century
BCE. See especially P. Hib I 73:9 where tto{tytov and bvoe are synonymous. The Egyptian
background of this word was also stressed by A. Wasserstein, “On Donkeys, Wine and the
Uses of Textual Criticism: Septuagintal Variants in Jewish Palestine,” in: A. Oppenheimer
and others (eds.), The Jews in the Hellenistic-Roman World, Studies in Memory of Menahem Stern
(Jerusalem 1996) 119*-142*, esp. 12*-129*. See further D. Wasserstein, “The Ptolemy and
the Hare: Dating an Old Story about the Translation of the Septuagint,” Scripta Classica
Israelica 17 (1998) 77-86.
See also Judg 19:3, 10: ayn - manuscript B: voc; manuscript A: UTogvytov.
20 CHAPTER ONE
thought that the translators replaced “1y1 with *p10n1, that is, ‘little
ones.’93 The same development took place in passage 11.
15. Lev 11:6(5)°4 Deut 14:7 MT NIN
LXX 8aovT0Sa = list ON NYE =O NYY
SacUTouc appropriately reflects n217x?° and therefore may reflect the
original Greek translation cited in the list by means of a literal
retranslation of its two elements. Rabbinic tradition emphasizes that
the translators avoided writing n2Ix in this place (that is, Aaywdv); it
may be that this claim reflects a post factum explanation; on the other
hand, \aywév may also represent the original translation of n2inx later
supplanted by SacvToéda.
33 See Aptowitzer, “Berichte” 2 (1909) 104-106; Geiger, Urschrift, 36; Talmon, “Scrolls,”
26; Miiller, “Nachrichten,” 81-83. This word is probably not originally Greek (Crys ; see,
e.g. the dictionary of Lewy) because it is hard to know what could be common to ‘11 and
the Greek word (inquirer), and since bio! is indeed attested in Hebrew (1QM, 11(7):3:
Onx¥a onins> wine X92 AW1 DION? AW) 2191) and in Aramaic (for the data, see S. Krauss,
Gnechische und Lateinische Lehnworter 1m Talmud, Midrash und Targum (Berlin 1899); see, e.g., T
in Cant 6:5 ]"p1Dx1). This word is apparently derived from Xb! (many manuscripts of b. Meg.
indeed read 1101 and not WwioNT as in the printed editions).
Even more has been written concerning *b10¥1 750, one of the three scrolls found in the
temple court according to y. Taanit 4:2, 68a; parallels in Abot de-R. Nathan, version B,
chapter 46; Sifre 356 on Deut 33:27; Sop. 6:4. See Talmon, “Scrolls” and the bibliography
there. It is unclear what was the character of Sefer Zaafute. In any case, the passage in
rabbinic literature apparently does not touch on passages 10 and 11 in the list. At most, it
may be claimed that these two passages do not belong to the list (note that they are lacking
in y. Meg. and in Mek.); however, such a claim is improbable in view of the parallel
between *b1pxT in the list and veavloxot in Exod 24:5.
4 The LXX to Leviticus changes the order of vv. 5 and 6. For the equivalent j5v -
AOL oy RUNS cf. Prov 30:26 (24:61) and Ps 104(103):18.
> The word is translated in the Syro-Hexapla of Deut 14:7 by x2inx (with SaouT0éa
added in the margin). See A. VGdbus, The Pentateuch in the Versions of the Syro-Hexapla, A
Facsimilie Edition of a Midyat MS. Discovered 1964 (CSCO 369; Louvain 1975).
CHAPTER TWO
THE FIFTH FASCICLE OF MARGOLIS’
THE BOOK OF JOSHUA IN GREEK
1. The four published fascicles
Margolis’ edition of the LXX of Joshua! was released between 1931 and
1938 (according to Jellicoe, SMS, 78), although the front page of all four
fascicles lists 1931 as the year of publication. The incomplete edition
(384 pages) contains the text of Josh 1:1 up to 19:38, ending in the middle
of a sentence.
This edition is unique, in so far as it contains a photomechanical
reproduction of a hand-written text. The contents, too, are unique.
Before Margolis, some attempts had been made to reconstruct the
original text of the LXX through an eclectic procedure,* but Margolis
tried to solidify this procedure by employing the principle established
by de Lagarde, viz., that of reconstructing the original text of the LXX
from its three major recensions.
As a first step, Margolis prepared a working copy of the text of these
three entities, viz., the Egyptian, Syrian, and Palestinian recensions.
The text of the central representatives of these three recensions was
printed in parallel columns, with the counterpart of MT printed
adjacent to the Greek text. L. Greenspoon had surmised that such a
manuscript existed,? so that its discovery (see below) came as no
surprise.
Initially, Margolis may have intended to publish his text of Joshua
as a three-column edition, but at a later stage he realized the
complexity of the textual transmission. In his critical edition,
IML. Margolis, The Book of Joshua in Greek according to the Critically Restored Text with an
Apparatus Containing the Variants of the Princrpal Recensions and of the Individual Witnesses, I-
IV (Publications of the Alexander Kohut Memorial Foundation; Librairie orientaliste Paul
Geuthner, Paris 1931 [- 1938]).
2 See especially P.A. de Lagarde, Genesis (Leipzig 1868); Psalms (Gottingen 1887); Judges
1-16 (Gottingen 1891); A. Rahlfs, Ruth (Stuttgart 1922); Genesis (Stuttgart 1926).
L. Greenspoon, “Max L. Margolis on the Complutensian Text of Joshua,” BIOSCS 12
(1979) 43-56, esp. SA.
De CHAPTER TWO
therefore, he represented four different recensions. In that edition,
Margolis did not present the text of the recensions as running texts, but
quoted individual readings from the reconstructed recensions in the first
apparatus whenever they differed from the reconstructed original
translation. In the Prefatory Note printed inside the cover the system
of notation is described as follows:
The Text as it appears on the top of the page is the nearest approach to the
Greek original as it left the hands of the translator(s). It has been arrived at
after a comparison of the remainders in the principal recensions, when once
the recensional peculiarities in each have been subtracted, and an
ascertainment of the form of the text to which the recensions lead and
which must be purified of the corruptions antecedent to them all. A
comparison of our most ancient manuscript (the Codex Vaticanus) with the
text here presented will show right in this first Part such conjectural
emendations as 3 16 (kapta@tav) 4 24 (katpw) 5 6 (TLV Souvat).
Below the Text is printed the Apparatus. It consists of (1) the variants
of the principal recensions: E S P.-C (M) (2) under the head of each of these
the evidence for its readings in the purer members and the defalcations on
the part of those given to mixture (impure members); (3) the variants within
the basic form of any recension; (4) marginal readings in the manuscripts
principally touching the later Greek translators ...
In this Prefatory Note Margolis remarked that ‘the full Introduction
will be issued with the last part,’ but it has never appeared. Since the
last part of the edition itself was lost in the 1930s (see n. 5), it has often
been surmised that the Introduction was also lost. Many introductory
remarks, however, were included in Margolis’ “Specimen of a New
Edition of the Greek Joshua,” Jewish Studies in Memory of Israel
Abrahams (New York 1927; repr. 1980) 307-323. In addition, Margolis
wrote some introductory studies which may have been intended as
sections of a larger Introduction.4 For the first sample of such an
introductory chapter, see Greenspoon’s study mentioned in n. 3. In the
meantime the edition itself remains our main source for understanding
the principles guiding Margolis in composing his eclectic text.
4A completed manuscript of this Introduction probably never existed, but good drafts
have been found, both by the present author in the aforementioned archive of Dropsie
College (1980) and by L. Greenspoon. See the latter, “A Preliminary Publication of Max
Leopold Margolis’s Andreas Masius, Together with His Discussion of Hexapla-Tetrapla,” in:
Salvesen, Origen’s Hexapla, 39-69. In n. 4 there Greenspoon mentions the preparations for
the publication of this Introduction.
FIFTH FASCICLE OF MARGOLIS’ THE BOOK OF JOSHUA IN GREEK 23
2. The missing part of the edition
The fourth fascicle of Margolis’ edition is incomplete as it ends in the
middle of 19:38. The original manuscript was probably lost in Europe.°
The missing part (pp. 385-475 of the book = fascicle V) was
discovered in 1980 by the present author in the archives of Dropsie
College (now: Center for Judaic Studies of the University of Pennsyl-
vania; previously: Annenberg Research Institute) in Philadelphia.®
This manuscript, though not the original which was probably lost, is an
excellent copy and has been published in 1992 according to Margolis’
original plan.” The photostat found in Philadelphia encompasses the
complete edition of Joshua. It contains several hand-written corrections,
probably inserted after the original manuscript had been sent to the
publisher.®
5 Jellicoe, SMS, 278: ‘The remainder of the manuscript (Part V and the Introduction)
must be numbered, it would seem, among the literary casualties of the Second World War,
since repeated inquires have failed to elicit any trace of it in Paris and it must be presumed
to have been irretrievably lost or destroyed.’ For a similar remark, see L. Greenspoon, “Ars
Scribendi: Max Margolis’ Paper ‘Preparing Scribe’s Copy in the Age of Manuscripts,’” JQR
71 sey) 139, n. 8.
Margolis taught for many years at Dropsie College. See H.M. Orlinsky, “Margolis’ Work
in the Septuagint,” Max Leopold Margolis, Scholar and Teacher (Philadelphia 1952) 35-44.
7 MLL. Margolis, The Book of Joshua in Greek, Part V: Joshua 19:39-24:33 (Monograph
Series, Annenberg Research Institute; Philadelphia 1992). xxvi + pp. 385-475, with a
preface by E. Tov.
In the following list, references are to pages and lines in the edition. Unless stated
otherwise, all details mentioned below replace elements in the edition:
98:12 au™ ]avTo.g yuan
104:17. Tar] add: yaw }
115:16 — tpovoyn] add: tpat&(n) FO
116:9 S.8pa xp |...F*...
116:14 = wevtxovta j...F"...
116:15 = odAKy...auToU F*
120:10 «at exavoav avta Fo
120:15 10F p.peyav
120:15 = ave [t]AaBev
120:16 axap Fo
122:11 aouha
135:18 omit: 69
143:17. Baotherg] prm kaToLkouvTEes &
145:16 = evJ> Fb
147:13. omit: ¢
162:15 e:nav Pd,
176:19 S:aBaceug F’d
188:19 avdpicecbe ] prm add
190:16 aut Javtous G:atavtac F’d
191:13 ek paxnda Pd
192:14 Katedovevoav ...F°¢ i
24 CHAPTER TWO
3. The system of the edition
Margolis’ editorial system is described briefly in his “Specimen.” A
full explanation of the editorial system was scheduled to appear in the
Introduction.
In the Prefatory Note to the edition, Margolis described the nature
of the edition and the grouping of the manuscripts as follows:
194:11 eEwdoOpevaav]..F...
196:20 aésoddAay | omit:
199:14 — ov(k)]...F...
208:12 vaded Swp | omit: >
208:13 = vayeoSup F*] add: vade808up
209:10 ~=—-1-4]..., 1 = PC b (= PC) S,
213:14 omit: (vid)
222:13. autouc (omit parentheses)
225:11 F#’™ Jadd:
225:17 — opouc J... Fd
231:6 2 (not: 1)
231:9 yepyeaoupt >
233:15 — rytov J...F*...
234:19 avtwv | autov
239:11 (o )apwé
259:20 wapaab > p
260:12 =a (not: a)
263:13 omit: dy
264:16 d.,at TodeLE avTwwy |...
285:10 Xaodwv ]j...d,...
309:9 end] add: S$, = P,C
359:14 = P,] add: 4n=§
365:21 =: 174,,] add axaceda6 a
367:10 uw0...ad,
373:10 BaGwp. a(a P)
In fascicle V the following corrections have been inserted:
424:20 BSretecev )....a...(not: a)
425:13. (adda ap)
433:16 6uivd np
437:15 086 086 KS auTOG...
445:14 emwvopacer ]j...(nt) ng
446:10 autov...d..] omit:
468:13 Tepeptvéov...a, (not: a)
469:15 kat etopevOnoav. ay..] omit: a,
471:6 Bupvaboapay...a ] omit: a
471:16 a=S@apvadbcoayxap n.™...
474:13, €avuTwy ToTOV...n
FIFTH FASCICLE OF MARGOLIS’ THE BOOK OF JOSHUA IN GREEK 25
4. Prefatory Note?
As the work is appearing in parts and as the full Introduction will be
issued with the last part, a prefatory note is in place to explain the
arrangement of the Text and Apparatus as well as the sigla. (...)
Below the Text is printed the Apparatus. It consists of (1) the variants
of the principal recensions: E $ P C (M); (2) under the head of each of these
the evidence of its readings in the purer members and the defalcations on the
part of those given to mixture (impure members); (3) the variants within the
basic form of any recension; (4) marginal readings in the manuscripts
principally touching the later Greek translators and other data concerning
them in so far as they have not found a place above. Lastly, brief notes
explanatory of the relation of the Greek to the Hebrew original and of
variants, recensional or intrarecensional, of the more difficult sort.
References to Masius are to a monograph on Syrus Masii being published
in the Harvard Theological Series.
The Sigla. Note that the wavy line under a letter indicates a recension
(e.g., E) and a circle around a large letter (e.g., (C)) marks a secondary
version and around a small letter (e.g.,(a)) a printed text.
1. E = Egyptian recension consisting of
O = Oxyrhynchus Pap. 1168 (a fragment = 4 23-5 1)
B = Vat. gr. 1209
B = Vat. gr. 1252
(B= Bohairic, lectionary (fragment)
© = Coptic (Sahidic) = © © (ed. Ciasca)
© ™ (ed. Maspero) © S (ed. Schleifer)
© (ed. Thompson)
e=S. Marci 4
€ (in certain parts) = Coisl. 3 (Paris)
(E) = Ethiopic = (©) (ed. Dillmann, specifically ms. F)
(EH (ms. Haverford College)
h (from p. 139 on) = Vat. Reg. gr. 1
Zz. 9 = Syrian recension consisting of 10
Oa = K (fragmentary) = Leipzig Univ. -Bibl. gr. 2
=ke-
® k; = Paris Nat. gr. 5
kz = Oxford Univ. Coll. 52
k = Moscow Syn. gr. 31
W = Wy = Paris Nat. gr. 6 and w2 = Athens Nat. 44
(L) = Old Latin (ed. Robert)
a ty = Vat. gr. 1901
to = Laur. Plut v 1
t=
t; = Laur. S.Marco 700
7 On page 5, I noticed the following misprint: instead of k=k=, read k= k=.
10 To these sigla, add $, = Greek members of S,
26 CHAPTER TWO
to = Paris Nat. er. 4
f = f = Ferrara Bibl. Com. 187 L. H and 188 II
f=
f; = Paris Nat. Suppl. gr. 600
fo = Zittau Stadtbibl. A. |
1 = Oxford Bod}. Laud. gr. 36
3) P = Palestinian recension consisting of
Py (Hexapla) =
G = Codex Sarravianus
y = Escorial Real Bibl. y-II-5
c=ce
cy = Vat. gr. 330
cy = Rome Chigi R. VI. 38
= Complutensian edition
= Lagarde’s edition
P> (Tetrapla) =
b = Brit. Mus. Curzon 66
(S)= Syriac =
(S)¥ = Brit. Mus. Lect. Add. 14, 485
a = Brit. Mus. Add. 12,133
M = Syrus Masii; Mas = the Greek of Masius
On = Ong and On, = Onomasticon Eusebii-
Hieronymi, ed. Klostermann
1 (parallel lection in 1)
4. C = Constantinopolitan recension consisting of
A = Codex Alexandrinus
M = Paris Nat. Coisl. 1
V = Vat. gr. 2106
W = Washington Smithsonian Freer Gall. |
a=S. Marci 3
v =v =5. Marci 6 and v9 = S. Marci 5
(a) = Aldina
g = Glasgow Univ. Libr. BE. 7°. 10
r = Paris Nat. gr. 1
v =S. Marci 2
€ (in certain parts), u from p. 231 on, h up to p. 139
(A) = Armenian
or M =a number of groups and single mss. which, while
not at all or (as the case may be) not necessarily
interdependent, rest for the greater part on C but admit
readings from the other recensions as well =
at
a= a, = Coisl. 2 and a = Paris Nat. gr. 2
a= a) = Laur. Medic. Palat. 242 and
ay = Vat. gr. 1657
F = Ambros. A. 147 inf. and ¢-Metewpa 461
d = d = Vat. Reg. gr. 10 and dz = Bodl. Canon. gr. 35
1 = Paris Nat. gr. 3
n (the Catenae group) =
FIFTH FASCICLE OF MARGOLIS’ THE BOOK OF JOSHUA IN GREEK 27
nj = Athos Aavpa 352
nz = Vat. gr. 2058
n3 = Athos Ilavtokpatopog 24
nq = Paris Nat. gr. 17 A
ns = Vienna Hofbibl. Theol. gr. 23
ng =S. Marci 15
nj = Laur. Acquisti 44
no = Vat. gr. 747
(N.B.n=nj1¢, 9 = 1, 2,Na=N2,5,Dp =N4, 3, 4)
p = Leningrad State Libr. gr.62 ~~
u =S. Sepulcri 2 from p. 1 to p. 280
R = Vat. Palat. gr. 431
6. Church Fathers
Eus(ebius)
Jus(tin)
Ory Or] = Origen
Thdt = Theodoret
Note the evidence for BewzlyMg A d2n2_-5nR has been taken from the
apparatus of the Larger Cambridge Septuagint.
Margolis’ description of the apparatuses is somewhat misleading, for
under ‘(3) the variants within the basic form of any recension,’ he
actually refers to two separate apparatuses. Apparatus 3 lists minuses
and contractions in certain witnesses, while apparatus 4 lists variants
within the recensional readings, which, according to Margolis had no
bearing on the main type.!!
In the Prefatory Note as well as in the apparatus itself individual
sources are neatly classified into five different groups. These comprise
four recensions (E, 5, P, C) and a group of mixed manuscripts (M). M does
not constitute a group in the same way as the first four, since it
represents a merely formal combination of mixed sources.!* In Margolis’
11 The first and main apparatus is relatively uncomplicated so that Margolis’ reasons for
accepting or rejecting readings can, as a rule, be easily reconstructed (see n. 26 below). On
the other hand, it is not always easy to follow Margolis’ decisions in apparatuses 2, 3, and 4.
Furthermore, the relation between apparatuses 34 and 2 is not always clear since they
overlap partially. Apparatus 4 is described as containing ‘variants within a recensional
reading which have no bearing on the main type,’ but they are often relevant to the main
type. As a result, apparatuses 1-2 now provide a seemingly unproblematical classification
of the evidence into four (five) recensions, as well as good background material for the
selection of the archetypal readings, but the really problematical evidence is often included
in apparatuses 3 and 4. For example, the evidence quoted in apparatus 4 often derives
from different recensions, so that new groupings are created. Furthermore, if the evidence
recorded for the individual recensions in apparatus 2 is arranged differently, it, too, goes
against the division into four (five) recensions (see, for example, the evidence relating to
So0A0¢ kuplou in 1:1).
12 Readings of M, which by definition are later than those of E, 5, P and C are recorded
in apparatuses 2-4 together with those of C.
28 CHAPTER TWO
words, ‘all of our witnesses are more or less mixed’ (“Specimen,” 308), so
that also other manuscripts outside the M group are recorded with
different recensions.
The division of the textual sources into four (five) groups reflects
Margolis’ view of the textual history of the LXX of Joshua, on which
the reconstruction of the original translation is based. This view
reflects the last stage of Margolis’ thinking, which underwent several
changes:
a. At first, when studying the transliterations of proper names in
Joshua,}8 Margolis distinguished between six different groups (‘mani-
puli’) of manuscripts,/4 arranged in two larger divisions (‘legiones’),
viz., a group centered around codex B (the [E]gyptian group = subgroups
bnh) and a group around codex A (the [P]alestino-Syrian group = sub-
groups oac). E is relatively free of Hexaplaric additions, and therefore.
presents the purest text. At this stage of the planning, Margolis wanted
to arrange the manuscripts in two separate columns.!>
B. At the second stage (1919), Margolis thought in terms of three
main recensions, viz., E,S and P. This stage is reflected in a mimeo-
graphed copy of the complete text of Joshua, found in the archive of the
Annenberg Research Institute. In this work, Margolis reproduced in
parallel columns the representatives of the three central recensions of
the LXX, with MT in a fourth column. This was a mere working copy,
not intended for publication, as is evident from the type of notations
written beside the text. The following manuscripts were chosen as the
13 “The Grouping of the Codices in the Greek Joshua,” JOR NS 1 (1910) 259-263. Also
other scholars—before and after Margolis—used the proper names in Joshua as a criterion
for classifying the manuscripts: J. Hollenberg, Der Charakter der alexandrinischen Ubersetzung
des Buches Joshua und thr textkritischer Werth (Berlin 1876); O. Pretzl, “Die griechischen
Handschriftengruppen im Buche Josue,” Bib 9 (1928) 377-427.
14 1) The Complutensian group (c): 19, 108, Compl.
2) The Aldine group (a) = 15.64; 18.128; Ald.
3) The Oxford (Arabian) group (0) = A, 29, 121 (82); N, 56, 71 (59); possibly M.
4) The Hesychian group (h): 44, 106; 54, 75, 118; 74, 76, 84, 106, 134.
5) The Catenae group (n): 16, 30, 52, 53, 57, 77, 85, 131, 144, 209, 236, 237, Cat. Nic.
6) The Sixtine group (b): B, 55, 63, 120, Sah., Eth., Cyr. Alex., Eusebius.
‘A critical edition of the Greek text of the Book of Joshua thus becomes a matter of
realization within sight ... The text should be printed in two columns corresponding to the
two forms which it assumed in Palestine and Syria on the one hand and in Egypt on the
other’ (ibid., pp. 261-262). Interestingly enough, at this stage of the planning, Margolis
rejected the type of edition that he would later seek to establish: ‘The tripartite reference
to Septuagintal transliterations in the current commentaries and lexica (to B, A, and
Lucian) is certainly convenient ... but is unscientific and should make way for a bipartite:
to post-Christian P, and pre-Christian E’ (p. 262).
FIFTH FASCICLE OF MARGOLIS’ THE BOOK OF JOSHUA IN GREEK 29
central representatives of the three recensions: codex B for E,!® codex b
for P,!” and codex k, for $.18 Between the columns Margolis added
variants pertaining to each recension, that is, differences between codex
B and the other representatives of E, codex b and the other represen-
tatives of P, and codex k; and the other representatives of S$
respectively. On the basis of these data, Margolis was able to compose
the archetypes of the three recensions, but it is not known whether he
ever prepared running texts of the archetypes. The final edition
(below) contains elements of the reconstructed recensions and no running
texts. 19
y. Continued research, especially of codex A, led Margolis to believe
that a fourth recension existed besides E,$, and P. The very existence
of C, a Constantinopolitan recension (with codex A as its major
representative) was first introduced briefly in 1925,29 and subsequently
explained at length in his “Specimen.”?!
The final edition (1931[- 1938]) reflects this third stage of Margolis’
thinking. In addition to the four major recensions, a fifth one (M) was
introduced. This group (Mixed texts) does not represent one of the
hyparchetypes, so that its readings were not listed in the first
apparatus. That apparatus lists only the readings of E, S, P, and C22
with separate listings of the subgroups of P, viz., P; (Hexapla) and P»
16 Margolis’ preference for B as a central text of E has been expressed in various places;
see especially “Specimen,” 316.
Bnit. Mus. Corzon 66 = x in Brooke-McLean, the main representative of P» (Tetrapla)
in Margolis’ edition. For Margolis’ preference for this manuscript within the P group, see
“Specimen,” 309; “Hexapla,” 136; “Ars Scribendi,” 147 (see n. 5 above).
8 Paris Bibl. Nat. gr. 5 = g in Brooke McLean = 54 in Holmes-Parsons. For Margolis’
preference for this manuscript within the $ group, see “K Text,” 3. In fact, the unical K
was pp ietereble, but that text is very fragmentary.
19 Such an edition would have tallied with de Lagarde’s principles, although Margolis
was probably less influenced by de Lagarde than is generally thought. Since Margolis
revised his views several times with regard to the number of recensions, his views were
probably influenced more by an internal analysis of the manuscripts than by any external
theory.
20 Margolis’ remarks are found in J.A. Montgomery, “The Hexaplaric Strata in the Greek
Texts of Daniel,” JBL 44 (1925) 298, n. 10. See further “Ars Scribendi,” 145. _
A similar division of the manuscripts of Joshua was suggested by G.V. Smith, An
Introduction to the Greek Manuscripts of Joshua: Their Classification, Characteristics and
Relationships, unpubl. dissertation, Dropsie College, Philadelphia 1973. Smith divided the
manuscripts into four families (Families 1~4) and he also characterized them in broad
terms. He relied heavily on the published work of Margolis and was not aware of the
a as sections found in the archives of what was then named the Dropsie College.
2 While it is unclear whether this sequence follows any principle, P and C are probably
juxtaposed because the latter depends on the former. The sequence may reflect the
relative closeness of the recensions to the presumed original translation, E being the closest
to that text, and C the most remote from it.
30 CHAPTER TWO
(Tetrapla). In some cases Margolis listed individual witnesses in the
first apparatus, especially when the combined evidence of these
witnesses transcends the borders of the individual recensions.”9
Also within this third stage of Margolis’ thinking developments are
visible, since the printed samples of chapters 6 (“Specimen”) and 174
differ in details from the final handwritten edition.2°
The notes in the edition explain the background of Margolis’
reconstruction of the recensions and of the original translation, but very
often the reader is left in the dark.
The central representative of each recension is remarkably close to
its reconstructed Urtext, that is, B to E, b to P, and kj, to §; likewise, E
(that is, basically codex B) is very close to the reconstructed Urtext of
the translation.2® Margolis only rarely allowed himself to deviate
from codex B, mainly in the transliterations of proper nouns.
In his edition of the LXX of Joshua, Margolis occupied himself
mainly with inner-translational problems. However, in the notes to the
edition he also remarked on translation technique and on the relation
between the reconstructed Urtext of the translation and MT.
23 This occurs especially with proper nouns, e.g., 10:33 Pa¢np; 12:7 AtBavou; 12:20
Mappwv; 12:23 Twetv. Since Margolis’ approach to proper nouns differed from that to
common nouns, the great majority of his conjectures refers to proper nouns. In fact, the
number of these conjectures is much larger than in any other edition of the LXX.
Found in the archives of the Annenberg Research Institute. The printing of this
chapter ts identical to that of chapter 6 published in “Specimen.” It is unknown whether
this chapter was once prepared for “Specimen” or derives from a stage when Margolis
wanted to have the whole edition printed rather than handwritten.
Minor differences between the “Specimen” and the “Prefatory Note” are also visible
in the grouping of the manuscripts. The latter represents Margolis’ final views.
Margolis indicated his preference for E in several places, e.g., “Specimen,” 316. Other
principles which guided Margolis in the choice of the ‘original reading’ are its remoteness
from the Hebrew and support by the majority of the recensions. F.C. Putnam, in his
seminar paper on “Margolis’s Textual Principles in “The Book of Joshua in Greek’ Based on a
Study of Chapters 8 and 19” (Dropsie College, 1981), reached the following conclusion: ‘In
narrative texts, as defined here, he generally chose the readings furthest from the Hebrew.
In the reconstruction of proper nouns, toponyms, he tended to choose the reading closest
to the Hebrew with some reservation for those readings which appeared to have been
revised or corrected toward the Hebrew.’
CHAPTER THREE
A COMPUTERIZED DATABASE FOR SEPTUAGINT RESEARCH
I. Nature of the database
1. Background
The CATSS project created a flexible multi-purpose database which
contains data needed for the study of the LXX and its relation to MT
(for bibliography, see section III). In the perusal of the database (see
section IV), certain types of information can be disregarded when
necessary, and other information can be added according to specific
needs.
The main section of the database is composed of the following
elements:
a. A parallel alignment of all elements of the MT and LXX. The text
of MT (BHS ) was encoded under the direction of R. Whitaker and H.
Van Dyke Parunak and verified by a team at the Westminster
Theological Seminary headed by A. Groves. The text of the LXX (the
edition of Rahlfs) was created by the Thesaurus Linguae Graecae in
Irvine, CA. The initial alignment of the LXX and MT was created in
1982-1983 by an automatic program written by J. Abercrombie, and the
results were corrected in accord with the project’s conception of the
equivalence of the MT and LXX by the team in Jerusalem. The
alignment of MT and the LXX creates exact equivalents of all elements
in both texts in two parallel columns:
1. Column a of the Hebrew records the formal equivalents of all
elements of the two texts, as if the LXX were a translation from MT. In
this recording, several types of symbols are used indicating special
phenomena and features which can be listed and analyzed separately.
2. Column b of the Hebrew records a selection of presumed equiv-
alents of the LXX retroverted from the Greek, when the Greek seems to
reflect a reading different from MT. It also records select differences
between the LXX and MT in the area of translational technique. The
main purpose of this column is to provide data which are not available
through the use of col. a.
32 CHAPTER THREE
B. The variant readings to the LXX (not yet integrated in the running
text of CATSS). The main Greek text incorporated in the database
follows the text of Rahlfs (to be changed to that of the Gottingen
editions when available), and to this text the full evidence of the
variants is added, either from the Géttingen editions or those of the
Cambridge series. For this purpose the contents of the apparatuses of
these editions are reformatted to the structure of the database, that is,
one Greek word per line. The variants are being encoded by the
Philadelphia team, and the system of recording the variants is
described by R.A. Kraft in CATSS 1.
y. A morphological analysis of all words in the LXX, that is, all
grammatical information relevant to the identification of the words,
including their dictionary forms (e.g., €pyopat for }A8ev). This includes
such information as the person, number, tense, mode, and type for verbs,
and the case, number, gender and declension for nouns. The initial
morphological analysis of the Greek words was produced with the aid
of an automatic program for morphological analysis of Greek, written
by David Packard and adapted for the LXX. The results of the
automatic analysis were verified and analyzed by the Philadelphia
team (see W.A. Adler in CATSS 1 and Textus 11 [1984] 125-139).
8. A Morphological analysis of all words in MT, that is, gramma-
tical information relating to all words in the Hebrew text.
Some forms of the CATSS database combine the morphological
analyses with the parallel alignment, while others do not.
2. Limitations of the database
The database does not provide answers to all questions in the study of
the LXX or of its relation to the underlying Hebrew text. It merely
contains data scholars would like to have available when analyzing
such issues. Some problems can be investigated only with the aid of a
computerized database. The flexibility of the database allows for the
inclusion of additional data at a later stage.
While most of the information in col. a is as objective as possible,
the recording in that column also entails subjective elements. Col. b is
fully subjective; yet, scholars will probably want access to this type of
material in spite of its subjective nature.
3. Nature and purpose of the Greek-Hebrew alignment
The philosophy of the alignment is to record as precisely as possible
the formal Greek-Hebrew equivalents of the LXX and MT. The relevant
COMPUTERIZED DATABASE 33
information is, as far as possible, contained in a single line of the
alignment with a minimum of cross-references to other lines, so that it
is easily accessed.
The basic principle followed in recording the equivalents is that of
formal representation (cf. TCU, 60-70). The formalistic approach
underlying the recording of the equivalents of the LXX and col. a of the
Hebrew implies that for the sake of argument the LXX is regarded as a
translation of MT. This is a mere convention adhered to by all biblical
scholars which promises the most objective basis for further research.
Yet, the procedure itself is problematic. Firstly, the LXX was not
translated from MT. Thus, in a book like Jeremiah it is unnatural to
record the details of the LXX as having been derived from MT, since the
LXX probably reflects an earlier stage in the development of the book
than MT (cf. Tov, “Jeremiah”*), even though on a technical level the
recording can be performed. Secondly, we do not know to what extent
the present eclectic editions of the LXX represent the original
translation. After all, the editions of Rahlfs and the Gottingen series
present mere reconstructions of that original translation. In spite of
these difficulties the margin of error for the reconstructions is probably
small (note the relatively minor differences between the Rahlfs and
Gottingen editions regarding the eclectic text, as opposed to their
apparatuses).
The main purpose of the alignment is to identify the Hebrew
elements which are equivalent with elements in the LXX, or, put
differently, which stand in the place of their counterparts in the LXX.
Necessarily, one often records Greek equivalents of Hebrew words
which differ from the words the translators had in mind or had in front
of them because of textual differences between the parent text of the
LXX and MT. These textual differences are referred to in col. b of the
database, but they are excluded from col. a which presents, as much as
possible, objective data.
According to this system, exegesis is disregarded in the notation.
Very free, paraphrastic, and even unusual renderings are recorded as
regular equivalents in col. a, since they reflect in some way their
counterpart in MT. For a detailed discussion of the problems connected
with recording the equivalents, see CATSS, vols. 1 and 2.
For an analysis of the determining of the equivalents, see part II.
4. Use of the database
In order to obtain the maximum amount of information from the
database, its various components can be merged for indexing and
34 CHAPTER THREE
compiling concordances. For these purposes the computer must be able to
make the link between words which are found in completely different
places in the alphabet, such as 727 and 72 in Hebrew. This
information is found in the aforementioned morphological analyses of
the Hebrew and Greek words.
One of the major reasons for creating a database is to enable easy
access to the data. These data can be stored in one form, and
reformatted in various ways, not only as running (consecutive) texts, but
also in other configurations. The data can be accessed in the following
ways:
a. Searches for individual words, combinations of words, or letter
patterns.
B. Indexing (‘sorting’) words in a particular part of the database or in
the database as a whole. Such an index can create a simple list of all
words in the exact form in which they occur in the text together with
all other information present in the same computer record (line). The
words can be sorted according to the desired alphabetical order (e.g.,
Hebrew, Greek). A similar index can be made on the basis of the
‘dictionary form’ (e.g., 127) in addition to the text form (e.g., 727).
y. Concordances. A concordance is based on the same principles as an
index, but it also supplies the context of the indexed word.
5. Special programs. Other information that is not easily available
through any of the three aforementioned formats can be obtained by
means of tailor-made programs.
With the aid of the computer, individual segments of the database
as well as the entire bank can be accessed in all these different ways.
New avenues are opened for the analysis of data in the realms of
textual criticism, language, and translation technique, as well as for
the study of all the corpora which depend on the LXX (see section II).
The data in the database can be accessed in various ways for word
searches and through indexes and concordances. The most sophisticated
program available so far is the Accordance program described in section
IV. Furthermore, various aspects of the translational technique
accessible through the database can now be researched. For some
examples, see Tov—Wright, “Literalness”*; G. Marquis, “Word Order”;
Nieuwoudt, Aspects, and see further section II]. Some details in the
notation may be singled out for analysis in wordprocessing programs.
The number in parentheses refers to the relevant paragraph in CATSS,
vol. 2.
a. Verses which the LXX has in excess of MT (4.2.1).
b. Asterized words in the LXX of Job (4.2.1).
COMPUTERIZED DATABASE 35
c. The Ketib-Qere variations in MT, including information on the
relation of the LXX to them (4.3.4, 60).
d. Prefixed and attached elements of the Hebrew words, with or
without their Greek equivalents, such as the prefixed -1, and the
prepositions -n, -2, -2 -2, and the pronominal suffixes (-1, -7, etc.)
(4.4.6). Cf. Tov—-Wright, “Literalness.”*
e. Differences in the numbering of verses between the MT and LXX,
often involving different text arrangements (4.5.5).
f. Representation of one Hebrew word by more than one Greek
‘main’ word (5.3.2.1).
g. Differences in sequence. The frequency of stylistic and gramma-
tical transpositions forms an indication of the literalness of the
translation (7.7).
h. Minuses and pluses of the LXX. Different categories of pluses are
distinguished (8.4.4) by F.H. Polak and G. Marquis, A Classified
Index of the Minuses of the Septuagint, Vol. I: Pentateuch, CATSS
Basic Tools, 4/1 (in press).
i. Doublets (10.1).
j. ‘Distributive’ renderings, that is, elements referring to more than
one word in the translation, such as pronouns, conjunctions and
prepositions (10.6).
k. ‘Repetitive’ renderings, that is, words occurring once in Hebrew,
and represented more than once in Greek (11.4).
1. Renderings of Hebrew prepositions by Greek compound verbs
(16.3.2).
m. Prepositions added in the LXX in accordance with the
translational habits of the various books (16.5.3).
n. Renderings of the construction %v? °n?0? (17.5.1). See Tov,
“Infinitive Absolute.”*
o. Transliterated Hebrew words (21.6).
p. Differences in verbs: active/passive (54.2.1.1).
q. Differences in prepositions (54.2.2.1).
r. Differences in vocalization (59.5).
s. Interchanges of consonants between MT and the presumed parent
text of the LXX, as well as metathesis and differences in word-
division (61.3). See Tov, “Interchanges.”*
36 CHAPTER THREE
II. Background of the representation of the equivalents
1. Formality
As a rule, Greek-Hebrew equivalents are determined easily as long as
one recognizes that formality is the overriding guiding principle
behind the notation. Thus all exegesis and possible textual differences
between the MT and LXX are disregarded in col. a:
Mich 4:5 ‘5 BTL
7s TAVTES
ony /7 ol Aaol
ioe TOpevOOVTAL
WN EKaOTOS
Y/7PX OW/2 Thy d8dv abtoi
Even though 1/0? ow/2 differs much from tThv d8dv abtod, the two
phrases are listed as equivalents which cannot be broken down into
smaller units.
Mich 1:5 “4/) Kal tls
ni 1) duaptta otkou
amily [ovda
The present notation demonstrates that it is hard to know whether
duaptia, ofkou (n°2 from nin2?), both, or neither reflect n2. It remains
true to say that nial and + duaptta olkou are formally equivalent.
The system of formal representation is not followed in all instances.
Occasionally that system is abandoned, and since the number of
exceptions is not very large and their nature can be formulated well, the
system itself is not harmed. Formal representation is abandoned when
it is misleading. The principle behind the formality is that the Greek
and Hebrew words stand on the same place and their listing as
equivalents is a good basis for further study, even if the Greek word
actually did not translate its Hebrew counterpart. However, the claim
that a Greek and a Hebrew word stand on the same place is misleading
when the Greek represents an element which is not present in MT and
when MT contains another word not represented in the LXX. The very
recognition of such a situation is to some extent subjective, and when in
doubt the formal approach is nevertheless applied. Thus, when
formality becomes misleading, that approach is abandoned, e.¢.:
Gen 4:25 ey a Eyuw 5
O1N Aday
TY —
+ 0X Evav
/ NWN TH yuvatka avTov
COMPUTERIZED DATABASE 37
Formality could require the listing of Evav as equivalent with 7Vy.
However, as long as 1)y is not considered graphically close to 7n (the
presumed equivalent of Evav), both 119 and Evav are recorded as having
zero-equivalents.
Deut 13:3 1371 TOpE VO LEV
“FINN —-
+ Kal AaTpEVOWLEV
On? Beots
ons ETEPOLS
On a formal level, *4nx and kal AaTpevowpev are equivalent. However,
AaTpevowpev has its real equivalent at the end of the sentence, so that
the above notation is more realistic.
Ruth 2:21 TNn/) Kal eltev
nn Povd
MPINVI/T a
7 TIpOS
+ Thy reve pav avTTis
On a formal! level, 7°281n/n is equivalent to 1pd¢ Thy TevOepav avdTis.
However, these words have completely different meanings and are
dissimilar graphically, so that both are denoted as having zero-
equivalents.
2. Split representation
The basis for the recording is either a single Hebrew word with all its
attached elements or two or more Hebrew words represented by one or
sometimes more than one Greek main word. As a rule, this system can be
followed conveniently, and complications are met when the Hebrew or
Greek word is represented by two or more words or parts of words which
are not consecutive. For these and other instances a procedure has been
devised for recording the information in such a way that all of it is
available at the stage of indexing. For this purpose details are
recorded twice, once in their actual place (for the sequence of the LXX
and MT is never abandoned) and once within special brackets ({}) in
accordance with their equivalents. All elements within these brackets
duplicate data found elsewhere in the text expressed as “...”:
Gen 3:23 {ow/n}... WRX éE tis
np? EAALGON
ow/n bayer
€E As reflects both wx and ow/n occurring later in the sentence.
38 CHAPTER THREE
Mich 4:3 {ayv...} 82/1 Kal oUKETL LL)
yan? pddwo1
abhi) {...}
Otxétt reflects both x? and 71y.
This system is also used for possible condensations:
Gen 7:11 niw/2 év {...}
MND wD Tu EEAKOOLOOTES
my {n3yw/3...} ETEL
niw/2 and mw are represented by one word only.
Gen 17:24 ]2 |
yens/) oyvwn évevnKovtTa évvéa
mw {72...} ETO
3. Inversion
The notation of inversions is problematic since the formal represen-
tation of the actual sequence would create unrealistic equivalents. At
the same time, there is no reason to deviate from the principles
described above, since also in other instances the formal representation
creates unrealistic equivalents. Furthermore, in many instances it is
unclear whether the LXX reflects the sequence of MT or an inverted one.
In the system of CATSS, unlike HR, whenever the LXX reflects a
sequence XY and MT yx, the equivalents are represented exactly as they
occur in the text, that is X-y, Y-x. In those cases, a reversal sign (~) is
used in the Hebrew column, after the first element and before the second
one. In some cases question marks are added after the reversal sign. The
real equivalents (presumed equivalents) are provided in col. b. This
system is used for the inversion of both adjacent and non-adjacent
elements referring to either one or more elements.
Gen 2:4 o-nw/) paX Tov ovpavdy Kal Thy yhv
recorded as onw =~ POX TOV OUpaVdV
Pux/) = onw/) ~ Kal THY yh
Deut 13:3 1373 TOpE VOD LEV
~ + Kal AaTPEVOWLEV
“INN —
DPN Bets
fabinlaly: ETEPOLS
TWN ols
NX? ovK
anv ol Sate
Osa a0I/ 1 =
COMPUTERIZED DATABASE 39
Grammatical and stylistic transpositions are treated differently from
the aforementioned system, since for them the actual sequence of the
LXX is not followed. In these cases the sequence of the LXX is
abandoned since these transpositions are part and parcel of the Greek
language. At the same time, when doubts arise with regard to the
choice between regular and stylistic transpositions, the former option is
chosen, since that system does not require the insertion of changes.
Grammatical and stylistic transpositions of the LXX are thus
represented in an inverted order; however, with the aid of the
aforementioned system of split representation which repeats infor-
mation, the actual sequence of the LXX is preserved. This system
applies especially to the post-position of particles.
Ruth 3:11 zis ae
yr ol ev
(Pa) Fas) yap
Gen 8:5 {en Tou S€KkaTou
wIn/7 {...~ Tou} unvdas
“w/t {...~ TOU S€KaTov}
4. Pluses and minuses
The LXX, which is the point of departure, contains both pluses (+) and
minuses (—) vis-a-vis MT.
The recording of pluses and minuses presents the most difficult
problems in the analysis as it is often hard to decide whether the
Greek word represents one or two words of the Hebrew or sometimes no
word at all. The subjective nature of the decision should be admitted.
Eg, Ruth 4:1 "22D —
"IN KpULE
The recording implies that kpvd.te represents only 71n?X (see the
variants) and not both Hebrew words. In a free translation unit and in
different circumstances, a different decision might have been made. In
cases of doubt both Hebrew words are listed as equivalents of the Greek
word.
Exod 32:34 ON els
+ TOV TOTIOV
WR Ov
*n4a7 eltrd
WD COL
40 CHAPTER THREE
According to one way of viewing the equivalents in this verse, els Tov
TOToV represents ?X. In a way, this is a formal presentation of the
evidence, since els Tov TOTOv represents the one Hebrew word found in
the slot between oyn and wx. However, there is also another way of
viewing the equivalence, and that, too, presents the data formally.
Since els represents ?X, it is not impossible that Tdv TdéTov reflects a
variant O1?n7, and hence it is preferable to list T6v TéTov as a plus (+).
This type of formal presentation is preferred, since only in this way
will the plus Tov téTov be available at the stage of indexing as a plus
element.
In order to improve the practicability of the index,’ —’ is listed also
for the lack of one out of two words written on the same line, especially
for nN:
Gen 36:6 —nNx/} Kal TdavtTa
W/ypn Ta UTdpxXovTa
5. Further details
The full description of the parallel Greek-Hebrew alignment involves
also the following features and words:
doublets (Greek and Hebrew)
Particles and conjunctions
the Greek article
ne
prepositions
elvat
numbers
words which have a double task in the translation
& Eu)
TUS
comparatives, superlatives
(2
WR
geographical names
autos
combinations of pronouns and verbs
combinations of conjunctions and verbs
additions of pronouns to nouns and verbs
special problems in the verb
infinitive absolute
differences in vocalization
COMPUTERIZED DATABASE 41
interchanges of noun + pronominal suffix / noun + noun
interchanges of noun / construct noun + noun
Ill. List of publications relating to the CATSS project
The bibliography is arranged chronologically and pertains to studies
describing and presenting the database as well as studies based on it.
R.A Kraft and E. Tov, “Computer-Assisted Tools for Septuagint
Studies,” BIOSCS 14 (1981) 22-40
J.R. Abercrombie, “Computer Assisted Alignment of the Greek and
Hebrew Biblical Texts—Programming Background,” Textus 11 (1984)
125-139
W. Adler, “Computer Assisted Morphological Analysis of the
Septuagint,” Textus 11 (1984) 1-16
P. Lippi, “The Use of the Computerized Data Base for the Study of
Septuagint Revisions,” BIOSCS 17 (1984) 48-62
Z. Talshir, First Esdras: Origin and Translation, unpubl. diss.,
Hebrew University, Jerusalem, 1984 (Heb. with Eng. summ.)
E. Tov, “The Use of a Computerized Data Base for Septuagint
Research—The Greek-Hebrew Parallel Alignment,” BIOSCS 17 (1984)
36-47
N. Leiter, “Assimilation and Dissimilation Techniques in the LXX of
the Book of Balaam,” Textus 12 (1985) 79-95
Tov—Wright, “Literalness’*
E. Tov, “Computer Assisted Alignment of the Greek-Hebrew
Equivalents of the Masoretic Text and the Septuagint,” in: N.
Fernandez Marcos (ed.), La Septuaginta en la investigacion contempor-
anea (V Congreso de la IOSCS) (Textos y Estudios “Cardenal Cisneros”
34; Madrid 1985) 221-242
B. G. Wright, “A Note on the Statistical Analysis of Septuagintal
Syntax,” JBL 104 (1985) 111-114
CATSS 1
W.T. Claassen, “Towards a Morphological Analysis of Biblical
Hebrew—A Semi-Automatic Approach,” Actes 1986, 143-154
J. Lust, “The Computer and the Hypothetic Translators of Ezekiel,”
Actes 1986, 265-274
G. Marquis, “Word Order as a Criterion for the Evaluation of
Translation Technique in the LXX and the Evaluation of Word-Order
Variants as Exemplified in LXX-Ezekiel,” Textus 13 (1986) 59-84
42 CHAPTER THREE
Z. Talshir, “Linguistic Development and the Evaluation of
Translation Technique in the Septuagint,” ScrHier 31 (Jerusalem 1986)
301-320
CATSS:2
Tov, “Translation Technique’”*
B.G. Wright, “The Quantitative Representation of Elements:
Evaluating ‘Literalism’ in the LXX”, in: Cox, VI Congress, 311-335
N. Leiter, “The Translator’s Hand in Transpositions? Notes on the
LXX of Genesis 31,” Textus 14 (1988) 105-130
R.A. Kraft, “Computer Assisted Identification and Reconstruction of
Fragmentary Manuscripts,” Proceedings of the Second International
Colloquium Bible and Computer: Methods, Tools, Results, Jérusalem, 9-
13 juin 1988 (Paris /Genéve 1989) 319-321
B.A. Nieuwoudt, “Computer Assisted Research of the Greek and
Hebrew Bible (II),” in: E. Talstra (ed.), Computer Assisted Analysis of
Biblical Texts (Amsterdam 1989) 101-118
id., “Beyond CATSS: Utilizing Relational Databases for Text-
critical Research,” Literary & Linguistic Computing 4.4, 254-259
id., “The CATSS Database: Progress in Research Procedures on Main
Frame and on Personal Computer,” Second Colloquium, 401-417
E. Tov, “Computer Assisted Research of the Greek and Hebrew
Bible,” in: E. Talstra (ed.), Computer Assisted Analysis of Biblical
Texts (Amsterdam 1989) 87-99
E. Tov, “Achievements and Trends in Computer-Assisted Biblical
Studies,” Second Colloquium, 33-60
B.G. Wright, No Small Difference—Sirach’s Relationship to its
Hebrew Parent Text (SCS 26; Atlanta, GA 1989)
T. Bergren, Fifth Ezra: The Text, Origin and Early History (SCS 25;
Atlanta 1989)
B.A. Taylor, “Evaluating Minority Variants within Families of
Greek Manuscripts,” BIOSCS 23 (1990) 31-38
Tov, “Infinitive Absolute”
DJD VIII
G. Marquis, “The CATSS-Base: Computer Assisted Tools for Septua-
gint Study for All—Transcript of a Demonstration,” in: Cox, VII
Congress, 165-203
E. Tov, “The CATSS Project—A Progress Report,” in: Cox, VII
Congress, 157-163
J. Lust, E. Eynikel, and K. Hauspie, A Greek-English Lexicon of the
Septuagint, I-IT (Stuttgart 1992, 1996)
COMPUTERIZED DATABASE 43
B.A. Nieuwoudt, Aspects of the Translation Technique of the
Septuagint: The Finite Verb in the Septuagint of Deuteronomy, unpubl.
diss. University of Stellenbosch, Stellenbosch 1992
F.H. Polak, “Statistics and Textual Filiation: the Case of 4QSam?/
LXX (with a Note on the Text of the Pentateuch),” in Manchester
Symposium, 215-276
Tov, “Interchanges”*
J.J.S. Weitenberg, Parallel Aligned Text and Bilingual Concordance
of the Armenian and Greek Versions of the Book of Jonah (Dutch
Studies in Armenian Language and Literature 1; Amsterdam 1992)
CATSS:3
B.A. Taylor, The Lucianic Manuscripts of 1 Reigns, Volume 1,
Majority Text, Volume 2, Analysis (HSM 50,51; Atlanta 1992, 1993)
Kyung-Rae Kim, Studies in the Relationship between the
Samaritan Pentateuch and the Septuagint, unpubl. diss., Hebrew
University, Jerusalem 1994
T. McLay, Translation Technique and Textual Studies in the Old
Greek and Theodotion Versions of Daniel, unpubl. diss., University of
Durham, 1994
B.A. Taylor, The Analytical Lexicon to the Septuagint—A Complete
Parsing Guide (Grand Rapids, MI 1994)
F.W. Knobloch, Hebrew Sounds in Greek Script: Transcriptions and
Related Phenomena in the Septuagint, with Special Focus on Genesis,
unpubl. diss., University of Pennsylvania (Philadelphia 1995)
F.H. Polak, “A Classified Index of the Minuses of the Septuagint,”
in: Greenspoon—Munnich, VIII Congress, 335-347
J.C. Treat, Lost Keys and Interpretation in Old Greek Song of Songs
and Its Earliest Manuscript Witnesses, unpubl. diss., University of
Pennsylvania, 1996
IV. Use of the database with the Accordance program
1. Background
The CATSS database, as well as the MT and LXX ‘text panes,’ can be
accessed with the aid of the Macintosh Accordance program,! as of 1998
without col. b of the Hebrew, and without the linkage with the CATSS
files of morphological analysis of the Greek and Hebrew words.
Nevertheless, the internal morphological analysis of Accordance
1 Thanks are expressed to Roy Brown, the programmer of Accordance, and to F. Polak for
improving the description.
44 CHAPTER THREE
allows the user access to many of the data which otherwise would
have been obtained by a linkage between the main file of parallel data
and the morphological analyses of the CATSS database. Complete
listings of individual Greek and Hebrew words can now be provided
with the aid of the internal Accordance predetermined lemmas
(morphologically and lexically tagged) and can be displayed with or
without the context of the verse. In this way all the individual words
of the Hebrew and Greek Bible can be concorded with their equivalents
in the other language. Furthermore, the grammatical analysis and the
search possibilities of Accordance allow bilingual grammatical
searches.
At the word level alone, the new type of concording retrieves much
more information than HR, as that tool does not include all the Hebrew
and Greek words. Thus, the user now has access to all the equivalents of
such Greek particles as 5€ and d\Ad and of all the Greek pronouns, and
in these cases the Hebrew parallel data are available as well. The
Accordance program further avoids the various pitfalls of HR’s
recording system (cf. TCU, 90-99), and it can execute searches of parts of
Hebrew or Greek words, such as Hebrew prefixes and suffixes and Greek
preverbs.* Beyond HR, Accordance enables searches of combinations of
words and of grammatical categories (see below). In the MT and LXX
‘text panes’ of Accordance (but not in the MT/LXxX file) searches can be
executed on any text unit in the LXX or the Hebrew Bible (all of the
LXX, one or more biblical books, or any combination of verses). Searches
can also be conducted on the comments in CATSS in the Greek and
Hebrew text relating to translation technique, the relation to the
Qumran scrolls, and underlying Hebrew variants.
Accordance furthermore provides the user with brief standard
equivalents (not always reliable) in English of all the words in the
Hebrew and Greek texts. This information is provided in the text files
by placing the pointer on the text word. The lexical box at the bottom of
the screen provides the Hebrew or Greek text word together with the
lemma word and its brief morphological analysis (thus by clicking on
nax71 in MT, the lexical box provides the different English equivalents
of 7x as well as their morphological analysis). More extensive lexical
information can be culled from entries in LSJ and the LXX lexicon of
Lust-Eynikel-Hauspie? for the Greek words and in BDB for the Hebrew
2 Words prefixed by -2 are searched for in the MT/LXX tool with the use of a ‘wild card’
according to the sequence of the Hebrew as: ? <within 2 words> 3.
J. Lust, E. Eynikel, and K. Hauspie, A Greek-English Lexicon of the Septuagint, I-U
(Stuttgart 1992, 1996).
COMPUTERIZED DATABASE 45
words. This information is provided by first selecting the word in MT,
and by subsequently selecting a lexical source (BDB, LSJ, or the LXX
lexicon) in the Amplify Palette, usually on the right side. As a rule,
the program makes the correct connection between the text word in the
running text of MT and the entries in BDB. Thus if Xx¥1n in Xx1n1 is
selected, the relevant entry of xx” in BDB is displayed. This search can
also be applied to the MT/LXX text, but as the Hebrew in that text is
not connected with an underlying morphological analysis, often the
wrong entry from BDB is displayed.
In Accordance, lexical searches can be executed on the Greek (LXX1)
or Hebrew (HMT) text panes separately or on the MT/LXX tool (=
CATSS). The principles guiding these searches in the text files are:
a. Words can be selected from the text and placed in the search box.
b. Words can be defined in the search box.
c. Words can be called up from the list of predetermined words, in
the Options box in the main menu, under Enter Lexical Forms (e.g. 1717) or
Inflected Forms (e.g., *n177).
d. Complex searches can be performed in the Construct window.
The principles for these searches are more or less identical when
searching in MT, LXX, or the combined MT/LXX tool, but in the latter
text (treated by Accordance as a tool, rather than a text) the options
are more limited as it is not linked with the list of predetermined
lemmas.
The following files may be opened:
a. the MT/LXX tool (the parallel alignment of CATSS without col.
b) by selecting the appropriate item from the New Window Palette,
usually on the upper right side. Alternatively this text can also be
opened by clicking on the ‘Open...’ item in the Edit menu.
b. MT (HMT), reflecting codex L.
c. the LXX (LXX1), reflecting the edition of Rahlfs.
d. any combination of these text panes, or a combination with one of
the English translations, SP, or the Vulgate. Within Accordance all
these texts are linked, so that they can be scrolled down together,
always showing the same verse in Hebrew/Greek, Hebrew/English,
MT/SP, etc. as the first item on the screen. Also dissimilar texts can be
linked with the ‘Tie To’ command. Any second text can be added to the
first one by selecting the appropriate file in the menu File, New Text
Pane (e.g., HMT + SP or HMT + MT/LXX) or by selecting them
separately. Subsequently the ‘Tie To’ command in the Windows menu
should be invoked in order to link these dissimilar files.
46 CHAPTER THREE
The texts are presented as complete verses, and not as individual words
as in the CATSS database. The combination of the MT (HMT) and LXX
(LXX1) text panes is very significant in the perusal of Accordance, as
the separate Hebrew and Greek files allow for more search
possibilities than the MT/LXX tool.
Beyond the general equivalents of verses in the MT and LXX text
panes, Accordance also provides equivalents at the level of individual
words (lines in the CATSS database), by using the MT/LXX tool. The
sophistication of Accordance allows the user many possibilities short of
a full morphological analysis, so that the lack of the CATSS
morphological analysis of Greek and Hebrew is felt less. Furthermore,
by using the “Tie To’ command, the HMT and MT/LXxX text panes can be
combined, so that the morphological analysis of the HMT text pane can
be used in order to show the complete MT/LXX contexts in the parallel
window. The same possibility also exists in the reverse direction: single
equivalents can be called up in the MT/LXX tool, while the full context
can be viewed in the parallel window in the text pane of either the
HMT or LXX.
2. Principles of search in the lemmatized Hebrew and Greek texts
i. Word searches
The principles of searching words or forms in the two types of text files
differ, because the text panes of MT (HMT in Accordance) and LXX
(LXX1) use predetermined lemmas (morphologically and lexically
tagged), while the MT/LXX tool is not lemmatized, and hence its
search options are limited.
Searches in the HMT and LXX1 text panes are executed by opening
these texts and by filling in the word in the search box as described
above (the Search mode itself is activated by first clicking on Mode). In
this search Hebrew vocalization and Greek accents are disregarded, so
that the results refer to the Hebrew consonants or Greek letters only. A
simple search thus often produces more items than asked for. This
limitation pertains to nouns, adjectives, and particles, and to a lesser
degree to verbs. However, by combining data from different sets of
information within Accordance, specific searches can nevertheless
usually be performed, with the exception of the search for some
homographs (Hebrew words belonging to the same grammatical
category, such as 127 and 723). Thus a search? for the three consonants
4 The search alphabet is based on the transcription alphabet of the CATSS database.
Thus in Hebrew & = A, =B,) =G, etc., and in Greek,a =A,8 =B,y =G, etc.
COMPUTERIZED DATABASE 47
723 can be accompanied by the definition Noun in the Tags menu (in this
case referring to both 123 and 727, but in most other cases referring only
to a single noun)? or Verb. In the case of Greek homographs, ev can be
defined as either Preposition (producing a list of occurrences of €v) or
Adjective (producing a list of occurrences of év). This amplified
description 1s obtained by combining the regular search with the
possibilities provided by the Tags menu. In this way tailor-made
searches can be conducted for specific verbs or nouns. Thus present tense
forms of \éyw can be searched for as \éyw®[VERB present]. The optimum
for this search is obtained by opening both the HMT and LXX1 text
panes (with the aid of the New Text Pane menu). This procedure
enables the scrolling down together of the two text panes. The
combination of these two text panes is needed, as the MT/LXX tool
cannot be used directly with the grammatical tags.
An example of a complex search of data which cannot be accessed
with the extant printed research tools is the search for any combination
of two or more words such as *) 9y (79 Dy).6
The following issues should be considered as well:
1. Searches for the Greek base forms, such as kvpLos for the noun or
épxoyat for the verb automatically list all inflected forms as well, thus
including kuplou, etc. for kUptos and such forms as éAevon and t\8ooav
for E€pxouat.
2. Searches can be conducted on any combination of Hebrew and Greek
characters, including ‘wild cards,’ as explained in the Accordance
manual. Wild cards for single characters are indicated by ‘?’. Thus in
the LXX text pane a search for ot? will produce listings for both 6tt and
Ste. Likewise, a wild card in the middle of word refers to any single
letter. Thus 7?3 will list any Hebrew word starting with a beth and
ending with a resh, with a single letter in the middle. The slash
separating between morphemes in the database itself (e.g., 7/9) is
disregarded in this search.’
3. The wild card * refers to any number of letters. The search of
*epxop* (with a star at both ends) provides all the inflected forms of
that verb, including preverbs (€€€pxyouar, teooépyopat, etc.), and inclu-
5 In this search, Accordance provides the results for the following items under 727,
which cannot be distinguished: word, plague, pasture, Debir, Debar.
6 This search yields the following results for the LXX: Sid Ajpatos (3 x), 8a
ee (8 x), Sta duvtis (10x), Ek (1x), Eml (6x), Em stdpatos (4x), éml ordpa
(1 x), éml TS otédpati (2 x), kabdtep (1 x), kata (3.x), kata td elpnéva (1 x), Kata ordua
as) weTd (1 x).
This implies, for example, that for this type of search */n-72 consists of four letters only
when the search refers to n?72 or 2°72.
48 CHAPTER THREE
ding inflected forms which have no consonants in common with epxou-,
such as dtredevon and Sir A8wpev.
4. In the Construct panel the same results can be obtained without
the use of stars: When the Greek text is displayed, one should select in
the File menu New Construct, Greek. LEX is placed in the bottom left
window, together with epyoyat from the list of lexical forms. The
search is then started after the two windows are first linked with
LINK in the Options menu under Enter Commands. In the same way all
infinitives of this verb can be listed by listing ‘infinitive’ in the space
under the Greek verb, or all non-infinitive forms by selecting the NON
box for the infinitives. In a similar way all occurrences of 7?7 can be
concorded in the Hebrew construct window, starting with the bottom
right window. Or, all forms of the type yon are concorded in the New
Construct window as a combination of the LEX form 2 and VERB, inf.
constr., combined by the command WITHIN 1-1 words.
5. Combined searches can be extended to more than one item
(commands: AND, NOT, FOLLOWED BY, PRECEDED BY, OR). Thus,
use of the AND command (Options: Commands) allows the user to find
all verses in which Aéyw and ktptos occur together, or all verses in
which forms of \éyw are immediately FOLLOWED BY kvptos. The
same pertains to more complex searches such as oUtws <WITHIN 2
Words> \éyw <WITHIN 2 Words> ktptos.
6. Secondary searches on the results of initial searches can be
executed with the aid of the CONTENTS command in the OPTIONS
menu. In the last mentioned example in paragraph 4, many equivalents
of the combination of 2 and the inf. constr. are provided, which can be
tabulated further. If from this list the equivalent 6tTt is singled out, the
following procedure needs to be followed after the initial results have
been concorded: another search menu needs to be opened (FILE, NEW,
SEARCH WINDOW). In this search window, write ‘6TL <AND>
[CONTENTS SEARCH],’ both to be selected from the OPTIONS,
COMMANDS.
it. Grammatical searches
Accordance includes an analysis of all the Greek and Hebrew words
defining each of these words grammatically (e.g., for "727: noun, plural,
masculine, construct). The program allows for a search of all the words
belonging to a specific grammatical category. Thus the user can ask for
all nouns, or more in detail, all plural nouns, or in still greater detail,
all plural masculine nouns, or all plural masculine construct nouns (such
as 1171), etc. These searches can be executed with or without the
COMPUTERIZED DATABASE 49
equivalents in the other language. After the Hebrew or Greek text is
chosen, the search can be performed on any of the grammatical
categories listed in the Tags menu. At a second stage the parallel
Hebrew or Greek text can be linked to the results of this search, so that
all the bilingual contexts are presented.
The Hebrew tagging allows, i.a., for a specialized search of forms
with a directional he, paragogic he (both under ‘suffix’ in the Tags
menu), infinitive absolute forms of the Hebrew verb, construct nouns,
dual forms of nouns, relative and interrogative pronouns, suffixes,
conjugations of the Hebrew verb, etc. Tagging of the Greek allows for
similar searches, such as a specific tense or aspect of the Greek verb.
Thus, the frequency of the aorist optative can be researched in this
way.
Special searches can be executed by combining specific Greek or
Hebrew words with grammatical categories. In the Construct window
these searches can be combined with various commands such as NOT
(under the word searched, not next to it), WITHIN, INTER, AGREE (all
in the central box). In this way one can list, for example, Tpogevyopat
FOLLOWED BY ‘Noun’ in order to examine the rectio of that verb. The
subjunctive forms of the verb not preceded by ov or p7 can be listed in this
way (Accordance User’s Guide, 9.8).8 The construction év + infinitive
(actually = év + To + inf.) can also be concorded in this way, with or
without elements intervening between év and tu (Accordance User’s
Guide, 9.9). The latter search is executed by writing év in the left box,
followed by ‘VERB, infinitive’ in the adjacent box and below WITHIN
(1-2 words) in the central box. In another instance, examples of Beds
without an article within five words before the noun are listed
(Accordance User’s Guide, 9.14).? By the same token all entries of 1x or
927 which are NOT verbs can be concorded.
3. Principles of search for the unlemmatized MT/LXX tool
Words in the unlemmatized MT/LXX tool (= CATSS) cannot be accessed
with the same sophistication as the separate LXX and MT text panes.
However, some simple searches can be executed by searching for strings
of letters in either language.
These searches are executed in the MT/LXX tool by locating equivalents
either in the context of a complete verse or as individual lines of the
8 Central box WITHIN, left bottom box ov, 1}, crossed out by NOT, and adjacent box:
VERB, subjunctive.
Central box WITHIN, left bottom box ‘art.’ crossed out by NOT, and adjacent box @ed¢.
50 CHAPTER THREE
CATSS database. For this purpose, in the MT/LXX tool, the box Entry is
opened and either ‘Hebrew’ or ‘Greek’ is selected in this box.
Subsequently a Hebrew or Greek word or combination of letters in the
text is selected and copied in the search box. Alternatively any
combination of letters can be written in the search box. Subsequently,
the results are displayed in lists of individual equivalents in the
MT/LXX tool or of lists of such equivalents within their context of the
complete verse. For the first line on each screen the text reference is
provided in the reference box. The results of the search for ?y includes
such forms as °/>y, 7°>/y, but not °ya or 2y1, so that the results form a
reasonable basis for research, although inferior to the results obtained
in the separate LXX or MT text panes. On the other hand, the results
include the parallel elements in the other language.
In the Hebrew text in the MT/LXX tool, the different morphemes are
separated by a slash. These slashes are treated as word separators, so
that a search for x12°/) should be phrased as
N72? <WITHIN 2 WORDS> 3.
In the MT/LXX tool, the results are listed for the database as a
whole, and cannot be subdivided into individual books of the Bible.
i. Special searches and notations
Special searches include an analysis of all paragraph divisions in the
MT text (© or 0) which can be searched in the HMT text pane.
In the CATSS database, special notations refer to select categories in
translation technique and other data. Searches of these notations can be
made on all the special notations in the Hebrew and Greek columns of
the CATSS database, relating to the translation character of the LXX
and its relation to the Qumran scrolls, as well as textual variations.
These searches can be executed in the MT/LXX tool, in the ‘Entry’ box
under Comments. A search for ‘?’ lists all these comments according to
the sequence of the text. Specific details which can be searched for
include:
C= {ec} conjectures in the Greek text
d={d} doublets
d:=(.3d} distributive use in the translation
P=) difference between MT and LXX in particle /
preposition
r= {7} element repeated in the translation
s = {s} superlative
t = {t) transcriptions
COMPUTERIZED DATABASE 51
nd, ad ,nad,v,etc. subdivisions in the renderings of inf. constr.
denoted as {!}nd, etc.
sp = <sp> agreement between the LXX and SP against MT
gq = <q4b> agreements with Qumran scrolls, in this case
relating to 4QNum>
yw interchange between yod (MT) and waw (LXX),
etc.
m metathesis
y- the LXX omits a yod
J the LXX joins two words
At this stage the following important components of the CATSS base
elements cannot be searched for: . { --- --+ !. The next release of
Accordance will address these issues.
ii. Special display
The Amplify Palette in the top right corner of the text panes allows for
special presentations of the search results:
1. The ‘Plot’ option provides a graphic chart displaying the results
of the search data according to book and chapter. For example, this
presentation enables the user to see graphically in which chapters in
Genesis the hiph’il forms of the verb are found, and in which chapters
and books in the Torah 77y occurs. The ‘Table’ option provides the
actual numbers of occurrences in each of the biblical books.
2. The ‘Analysis’ box, to be used in conjunction with the ‘Analysis
display’ in the Options menu, lists the individual searches alpha-
betically, a feature which is of help in grammatical analyses.
3. Under ‘Parsing’ the morphological information for each of the
words is provided.
4. The box ‘Old Testament’ provides parallels from the books of the
Hebrew Bible, if extant.
5. The box ‘Speech’ activates the speech representation of any
element on the screen, in English, Greek, or Hebrew, including the
recitation of the complete Bible text, or the parallel alignment of
CATSS;
CHAPTER FOUR
GLOSSES, INTERPOLATIONS, AND OTHER TYPES OF SCRIBAL
ADDITIONS IN THE TEXT OF THE HEBREW BIBLE
I. Glosses in Greek and Latin texts from antiquity and the Middle Ages!
The appearance of glosses, interpolations, and other scribal additions
in the textual traditions of the Hebrew Bible has special relevance to
the LXX. First some definitions relating to different types of additions
to the text:
(1) Glosses. Ancient and medieval manuscripts contained many
glosses, as defined by the usage of the word in Latin and not according to
the original meaning of the word in Greek. In the study of ancient Greek
and Latin texts the term ‘glossa’ carries a very distinct technical sense,
which is also applied to medieval texts, though with some differ-
ences:* ‘A word inserted between the lines or in the margin as an
1 Bibliography: Delitzsch, Lese- und Schreibfehler, 132-143; Dijkstra, “Glosses”; Driver,
“Glosses”; M. Elyoenay (Kantrowitz), “Explanations to Ancient Words of Difficult Meaning
in the Text of the Bible,” in: Hagut Ivrit b©Europa (Tel Aviv 1969) 41-48 (Heb.); M.
Fishbane, Biblical Interpretation in Ancient Israel (Oxford 1985) 38-43, 166-170; Fohrer,
‘“Ezechiel”; Freedy, “Ezekiel”; Hall, Companion; J. Herrmann, ‘‘Stichwortglossen im Buche
Ezechiel,”” OLZ 11 (1908) 280-282; idem, “’Stichwortglossen im Alten Testament,’” OLZ 14
(1911) 200-204; R.W. Klein, Textual Criticism of the Old Testament (Philadelphia 1974) 32-36; J.
Krecher, “Glossen,”’ Reallexikon der Assyriologie und vorderastatischen Archdaologie I (Berlin/
New York 1957-1971) 431-440; Kutscher, Language; McCarter, Textual Criticism, 32-36; L.D.
Reynolds-N.G. Wilson, Scribes and Scholars—A Guide to the Transmission of Greek and Latin
Literature (3d ed.; Oxford 1991); P. Rost, ‘‘Miscellen, I. Ein Schreibgebrauch bei den
Sopherim und seine Bedeutung fiir die alttestamentliche Textkritik,”” OLZ 6 (1903) 403-407,
443-446; 7 (1904) 390-393, 479-483; S. Talmon, ‘Aspects of the Textual Transmission of the
Bible in the Light of Qumran Manuscripts,” Textus 4 (1964) 95-132 = Cross-Talmon, QHBT,
226-263; J. Weingreen, ‘’Rabbinic-Type Glosses in the Old Testament,” JSS 2 (1957) 149-162.
4 Oxford English Dictionary (2nd ed.; Oxford 1989) VI, 591. The Dictionary adds: ‘hence
applied to a similar explanatory rendering of a word given in a glossary or dictionary. Also,
in a wider sense, a comment, explanation, interpretation.’ The Oxford Classical Dictionary
(2d ed.; Oxford 1970) [= OCD] subdivides the entry ‘glossa’ into two sub-entries, focusing
on the meaning of the word in respectively Greek and Latin sources. For the former OCD
provides the following definition:
54 CHAPTER FOUR
explanatory equivalent of a foreign or otherwise difficult word in the
text.’3 Of a different nature are glosses in Sumerian and Akkadian texts
(see below), since these glosses, often written within the text, were
meant to be an integral part of that text.
(2) Scholia. Like the gloss, the scholion (ox6ALov) is a marginal
note, but usually it explains a difficulty in the context or section rather
than the meaning of individual difficult words. Individual scholia
usually are part of a larger collection of scholia on the text, forming a
commentary to that text. Thus, while the term gloss is usually applied
to sporadic interpretations of separate words, the scholion is part of a
hermeneutic tradition. Some scholia are short, similar in length to
glosses, while others are more extensive.’ In the world of the Bible the
Masoretic notes are the closest parallel to the scholia from the Greek
and Latin world. In classical Greek and Latin manuscripts glosses and
schoha usually are not found together in the same manuscript.
Glosses and scholia were not the only elements that were added to
texts upon the completion of the copying. Six additional types of
elements were sometimes added as well.
(3) Interlinear and marginal corrections. Single letters or complete
words were inserted into the text by the original scribes or subsequent
ones after the copying of the main body of the text was completed when
it was recognized that the text was wrongly copied. In principle,
correctional activity could also adapt the manuscript to another
manuscript of the same composition, but there is little evidence for such
practices in the case of the Bible.
(4) Exegetical elements (‘interpolations’). Such elements were often
inserted into the text by later scribes or readers, in the margin, between
In Greek literary criticism yAdiooat meant any words or expressions (not being
neologisms or metaphors) & ov&els dv elmo év TH Siadéxty (Arist. Poet. 145832), i.e.
belonging not to the spoken language familiar to the critic (14586), but to a dialect,
literary or vernacular, of another region or period (1457°4),
The modern use of the term ‘gloss’ does not reflect the meaning of that word in Greek,
but rather that of the identical word in Latin, described as following in the same
Dictionary: ... marginal or interlinear interpretations of difficult or obsolete words.
3 At a second stage these glosses were often collected, alphabetically or not, as so-called
‘glossae collectae’ or glossaries, and some of these actually constitute the basis of primitive
dictionaries of equivalents. These glossaries were numerous in antiquity, and even more so
in the Middle Ages. See especially the detailed descnption by B. Smalley, The Study of the
Bible in the Middle Ages (3d ed.; Oxford 1983) 46-66. In fact, although the glossaries had their
Origin in the margins of manuscripts, once collected, they became independent units and
thus started a life of their own.
See the descriptions, definitions, and bibliography in the relevant entries of the OCD
and apud B.M. Metzger, Manuscripts of the Greek Bible (New York/Oxford 1981) 46-48.
GLOSSES 55
the lines, or even in the text itself (by the original scribe, copying from
an earlier manuscript).
(5) Variant readings. Deviating textual traditions, deriving from
external sources (additional manuscripts of the same composition) and
relating to readings included in the body of the text, were sometimes
recorded in the margin, or, in Sumerian and Akkadian texts, in the text
itself, in the latter case often separated by a special sign.” Modern
readers used to critical apparatuses are too quick in considering
marginal notes as variant readings; however, the notation of variant
readings was probably very rare in antiquity. In the case of the biblical
text, it is not impossible that many of the readings preserved by the
Masoretes as Qere originally constituted such variants.
(6) Scribal remarks and signs. Marginal notes reflecting scribal
remarks are rare in some texts, but more frequent in others. They may
draw attention to a detail in the text, or introduce an addition to the
text, or remark on the condition of the material (thus in Akkadian
texts—see Krecher [see n. 1], 438-439). Scribal signs (not remarks) of
different types are not infrequent in some Alexandrian Greek texts and
several Qumran texts. They refer to such matters as new paragraphs
and elements in the text to which attention is drawn.
(7) Remarks on the content. These, too, were probably rare. MT TS V
(not the LXX) contain such a note in the body of the text.®
(8) Headings to sections of the text.”
It is not easy to distinguish between these eight groups of additions to
the body of the original text, partly because the borders between the
types of additions are often not well defined, partly because manuscript
evidence about the first stage of the addition is usually lacking. The
purpose of these groups of additions is different, so that they should be
kept separate as much as possible. Of these, some groups tend to be
written especially in the margin and others between the lines, but with
the lack of evidence on the original documents no clear statements can
be made. In certain manuscripts scholia or glosses are very frequent. In
other manuscripts corrections are frequent. The interlinear and
5 The inclusion of the additional elements in these texts was executed in a variety of
ways. Sometimes it was written in smal] signs next to the word it referred to, at other times
it appeared between that word’s different components, or was written at the edge of the
tablet, and at again other times it was separated from the preceding word by a special sign.
That sign, named ‘Glossenkeil,’ appeared in different shapes, among them a diagonal line
and a double-wedge shape. On all these systems, see Krecher (see n. 1) 433.
Jer 51:64 ‘Thus far the words of Jeremiah’ (the next chapter serves an appendix to the
book).
7 In the biblical text, see, for example, Jer 23:9 ‘Concerning the prophets.’
56 CHAPTER FOUR
marginal addition of exegetical additions (interpolations), scribal
remarks, remarks on the content, headings, and variant readings is rare
in the known manuscripts. The addition of marginal glosses and scholia
was institutionalized, so to speak, since they were often transmitted as
such with the manuscripts, while interpolations and corrections were
not preserved as such, since they were meant to be included in the body
of the text during the next phase of the text’s transmission.
For textual critics all these groups of additions to the text are
noteworthy since they show the multifaceted textual transmission in
all its complexity. Furthermore, these additions are of particular
interest, though also a source of frustration, since often they create
much confusion when texts containing these additions were copied into
new copies. For while elements written in the margin could be
separated from the textual transmission of the body of the text,
mishaps were bound to occur with interlinear additions, whatever
their original intention may have been. The basic problem for copyists
of texts containing any type of additions is how to copy these added
elements, if at all. This problem pertained in particular to interlinear
additions. Scribes who added the elements between the lines wanted
some of them to be part of the running text (corrections and
interpolations), while other elements were not meant to be made part of
the running text. Hence, by definition any scribe who at a second stage
inserted earlier glosses, remarks on the content, scribal remarks, or
variant readings into the body of the text, as it were, took a wrong
decision; in other words, by acting contrary to the intention of the
person who added a certain element, he created an unnatural or corrupt
text. Many an addition thus entered the running text wrongly.
We do not know how specific texts developed and therefore most
assumed additions to the original texts are only hypothetical. Unfor-
tunately, scholars are rather quick in assuming that words in the body
of the text derived from such interlinear elements incorrectly
integrated into the text. As a matter of fact, Reynolds and Wilson (see
n. 1), 206 assume that only infrequently were glosses incorporated in the
body of Greek and Latin texts. In our view, a similar understanding
pertains to the transmission of the Hebrew Bible. There are no clear
statements on this phenomenon in the scholarly literature, but often the
impression is created that glosses and interpolations were frequently
inserted into the text.
We now turn to the presumably incorrect integration into the text of
elements that were not meant to be there, in the first place of glosses.
At this stage of the discussion interpolations and corrections are
GLOSSES 57
disregarded, since they were meant to be an integral part of the text.
The other types of additional elements, such as scribal notes and
remarks on the content as well as variant readings were rare and they
therefore concern us less. Scholia were usually written in the margin,
and they were not confused with the running text, and besides, since
there are no scholia on Hebrew biblical texts, they need not be treated
here. This leaves the interlinear glosses as the major source for
confusion.
The assumption that interlinear additions were at a second stage
incorporated into the running text is supported by explicit manuscript
evidence from Greek and Latin texts (see Reynolds-Wilson (see n. 1), 207
for an example). Such direct evidence is, however, extremely rare. Less
rare, but still rare, are actual interlinear additions of glosses in Greek
and Latin manuscripts from antiquity. Admittedly for most of the
presumed glosses no manuscript evidence is available, since these
glosses presumably had already been integrated in the text itself. The
recognition of such glosses is mainly based on content analysis when
scholars recognize that a given word which could be taken as an
explanation of another word in the context does not suit the syntax or
context. In such cases it is often suggested that that word, originally
serving as an interlinear gloss, had been wrongly inserted into the text.
Strictly speaking, the element which has been recognized as
constituting a misplaced gloss must, according to the scholar, be
removed from the text, and should, in his mind, be reinstated in its
proper position as an interlinear gloss in a manuscript. The assumption
of some such glosses is supported (not proven) by their absence in certain
textual witnesses. If such support is altogether missing, the assumption
of a gloss constitutes a special type of emendation, for which content
analysis is our only guide.
Misplaced glosses are more easily detected in poetic texts than in
prose because of the structure of poetry. For a large collection of
examples for classical Greek and Latin texts, see Hall, Companion, 193-
197 and see further Reynolds-Wilson (see n. 1), 206.
II. Marginal and interlinear glosses in the manuscripts of the Hebrew
Bible?
When turning to the textual transmission of the Hebrew Bible, we note
that more or less the same types of scribal additions as have been
recognized in Greek and Latin as well as in Sumerian and Akkadian
texts have been surmised for the early manuscripts of the Bible. In the
58 CHAPTER FOUR
history of research this is not surprising, since many categories of
thinking have been transferred from the study of the classical texts to
that of the Bible. Each case of transferring patterns of thinking from
one area to another should be evaluated in its own right, so that
generalizing conclusions concerning the value of a comparative
procedure should be avoided. In favor of transferring assumptions from
the Greek scribal tradition to that of the Bible it should be noted that
many parallels in details have been recognized between the two
(especially in scribal signs and habits). However, we will see below
that the case of glossing is different, so that the existence of parallel
procedures actually has not been proven.
At this stage a brief remark is in order on terminology. The
terminology used in biblical scholarship with regard to scribal
additions is less precise than in classical research. While there are
differences between individual scholars, most of them indiscriminately
use the term ‘gloss’ for all types of added elements and this situation
complicates our analysis. Especially confusing is the habit to use the
term gloss for exegetical additions (interpolations). As mentioned in n.
2, glosses are basically ‘. . . marginal or interlinear interpretations of
difficult or obsolete words,’ while interpolations are (exegetical)
elements added to the text, explaining the base text or changing its
implication. A basic distinction between these two types of additions is
that an interpolation (exegetical addition) was meant by the scribe to
be part of the running text, while a gloss was not.
One of the difficulties in the analysis of assumed glosses and
interpolations lies in the lack of convincing evidence in the ancient
sources. For, while in classical and Mesopotamian texts we can actually
point to the physical existence in manuscripts of various types of added
elements, only limited evidence is available for biblical texts. In the
period preceding the discovery of the Qumran scrolls, it seems that no
marginal glosses (or, for that matter, any marginal notations) were
known at all from Hebrew biblical manuscripts, and it seems that this
claim still holds today for glosses, even after the discoveries at
Qumran. Nevertheless, the assumption of misplaced glosses and other
elements in the running text was and is widespread in biblical studies.
8 This difficulty is apparent in the terminology used by Fohrer, ‘“Ezechiel” and Delitzsch,
Lese- und Schreibfehler (for whom some glosses are ‘clarifying’ and others ‘correcting’ and
‘cross-referring’). It is not impossible that this terminology derives from the scholarly
literature on the Sumerian-Akkadian texts. Every element appearing after the ‘Glossenkeil’
(cf. n. 5) in that literature is named a gloss, not only glosses proper (in this case, translations
of words), but also phonetic instructions, textual variants, and scribal remarks.
GLOSSES 59
Possibly some scholars considered the marginal Qere a good parallel
for the assumption of glosses. But the nature of these Qere notes differs
from the assumed glosses. For almost all the assumed glosses are
exegetical, while the Qere notes are not. The latter notes should be
considered as variant readings, corrections, or a combination of the two
practices.? Another type of supporting evidence invoked before the
discovery of the Qumran scrolls was that of the ancient versions. Their
evidence could be relevant when elements of MT are lacking in one of
the versions (especially the LXX), or conversely, when elements found
in one of the versions are lacking in MT. But the data in the versions are
of a different nature, and they do not necessarily constitute relevant
‘evidence.’ When a word suspected as a “gloss’ in one source is lacking in
another textual witness, its very absence may support the assumption of
a gloss, but does not prove it. After all, anything could have happened
to the texts in question, including the omission or addition by the
translator, without any connection with the phenomenon of glossing.
It was thus premature to claim before 1947 that the biblical text
contains glosses or other types of scribal additions, and in our view this
situation has not been changed with the discovery of the Qumran
scrolls. At first sight it would seem that these scrolls finally provided
the factual basis for the assumption of extraneous textual material in
the margins and between the lines, and this type of argument is indeed
used in the literature, especially by Dijkstra, “Glosses.” A mere glance
in 1QIsa? shows several hundred elements added between the lines and
a few added in the margin. It was, however, recognized that this
material is not relevant to the issue of glosses in the biblical text. For
with one or two possible exceptions (see below), the additional
elements in the scrolls contain no glosses.!9 They also do not contain
scribal remarks (as distinct from scribal signs found in several texts),
headings, remarks on content, or scholia. This then reduces the possible
explanations of the additional elements in the Qumran manuscripts to
three: corrections, exegetical additions (interpolations), and variant
readings. There is little evidence in favor of the second and third
options, so that we are left with the first one. Indeed, the scrolls
contain many instances of obvious scribal correction, reflected by
cancellation dots, the reshaping of letters, and crossing out with a
9 For a discussion of these possibilities, see TCHB, 58-63.
10 See EF. Tov, “The Textual Base of the Corrections in the Biblical Texts Found in
Qumran,” in: D. Dimant and U. Rappaport (eds.), The Dead Sea Scrolls—Forty Years of
Research (Leiden/Jerusalem 1992) 299-314.
60 CHAPTER FOUR
horizontal line.!! Furthermore, the great majority of the additional
elements agree with MT (and the ancient translations when their
evidence is relevant).!2 The combined evidence of the correcting
techniques and an analysis of the content of the added elements leads us
to the conclusion that these additional elements are in the nature of
corrections. The exact pattern of the agreements of these corrections is
somewhat unclear. The agreement with MT is misleading, since the
added elements usually agree also with the other textual witnesses.
These corrections thus do not reflect a consistent revision toward MT, but
they probably adapted the text written by the first scribe either to the
base text from which it was copied or to a text used by a later scribe, or
to both. In any event, the text to which the copied text was corrected
happened to agree with one of the proto-Masoretic texts.
But even if there remains some lack of clarity regarding the nature of
the correcting process, one thing is clear. The great majority of the
added elements in the Qumran scrolls are neither glosses as defined
above nor interpolations (exegetical additions). In the literature the
term ‘correcting gloss’ is sometimes used, but this term of compromise
makes the terminology unnecessarily vague. Glosses and corrections are
different categories for scribes. The great majority of the added
elements in the scrolls can only be viewed as corrections. Some of the
elements added in the scrolls have been left out from the text originally
copied, and as a rule the text is erroneous or does not read smoothly
without them.!$
Direct evidence for the existence of glosses is lacking for manuscripts
of the Hebrew Bible, but the assumption that glosses were once
incorporated in biblical manuscripts is supported by the occurrence of
different types of marginal and interlinear additions in the following
sources: several Sumerian and Akkadian texts,!4 among them the
Amarna letters,!° a Ugaritic text,46 many Greek and Latin texts from
11 For some examples, see Tov, TCHB, 213-215.
12 For 1QIsa# see Kutscher, Language, 522-536. For a scroll like 4QJer*, which contains
many corrections, the evidence is overwhelming (see DJD XV, 151-154).
Isa 20:2 dyn P82" nanos; 23:8 aw TU spe: 65:12 anmina "815, and elsewhere.
14 See n. 5 above. For a very detailed description of the different types of glosses in this
literature, see Krecher (see n. 1).
See F.M.Th. BGhl, Die Sprache der Amarnabriefe mit besonderer Beriicksichtigung der
Kanaanismen (Leipziger Semitistische Studien V,2; Leipzig 1909) esp. 80-89; P. Artzi, ‘““The
Glosses in the El-Amarna Tablets,”” Bar Ilan Annual 1 (1963) 24-57 (Heb.); Krecher (see n. 1).
16 See S.E. Loewenstamm, ‘Eine Lehrhafte Ugaritische Trinkburleske,” UF 1 (1969) 74.
GLOSSES 61
antiquity,!” TN, various manuscripts of the LXX and V,!8 as well as a
rich scribal tradition from the Middle Ages in all languages,!?
including Hebrew sources.29 Only one such example is known from the
Qumran texts.
Isa 7:24 MT mwianaw (= LXX TS V)
thornbush and thistle
1QIsa2 mwi-rny 772
iON thornbush and thistle (the addition is
interlinear, above nw).
In this case the added word in 1QIsa® explains a word in the text.2!
In the same way as the Qumran scrolls reflect no glosses, they
contain no, or hardly any, interlinear or marginal interpolations
(exegetical additions).
An example of a possible (grammatical) interpolation is the
following instance:
Isa 44:3 MT 4y7t Dy cnn px Awa 2y oO? NOX IV ON PEN 7D
Even as | pour water on thirsty soil, and rain upon dry
ground, <so> will I pour my spirit on your offspring.
1QIsa4 vay BY on PRX Pawar Dy O21) ROY PV OM PEN ND
The word ‘so,’ added in modern translations, was also added in
1QI[sa?.
The absence or rarity of physically recognizable interpolations does not
exclude the possibility that some of the plus elements of the scrolls vis-
a-vis the other texts are actual interpolations that had been
exegetically added by scribes in the body of the text. These assumed
interpolations, however, do not establish the scribal habit of adding
interlinear or marginal interpolations.
We now summarize our analysis of the ancient evidence. Even if
hardly any glosses can be located in the known textual witnesses, it
does not necessarily follow that the ancient Hebrew texts were not
17 Fora large collection of examples, see Hall, Companion, 193-197; see further Reynolds-
Wilson (see n. 1), 206.
8 See C. Morano Rodriguez, Glosas Marginales de Vetus Latina en las Biblias Vulgatas
Espariolas (Textos y Estudios ‘Cardenal Cisneros’ 48; Madrid 1989).
19 See, i.a., B. Smalley, ‘’Glossa ordinaria,”” Theologische Realenzyklopadie XII (Berlin/New
York 1984) 452-457.
0 For Ben-Sira, see W. Caspari, ‘Uber die Textpflege, nach den hebraischen Hand-
schriften des Sira,’”” ZAW 50 (1932) 160-168; 51 (1933) 140-150.
1 In the spoken language of the Second Temple period, nv had a secondary meaning
of ‘iron,’ to which the glossator probably referred. Relevant material was collected by S.
Lieberman, ‘‘Forgotten Meanings,’” Leshonenu 32 (1967-1968) 99-102 (Heb.); E. Qimron,
Textual Remarks on 1QIs4,” Textus 12 (1985) 0-vs (Heb. with Eng. summ.).
62 CHAPTER FOUR
glossed. For one thing, the absence of glosses in the Qumran texts should
not be taken as representative for the transmission of the biblical text
as a whole since the Qumran scrolls are relatively late in the
development of the biblical text. The biblical text could still have been
glossed at an earlier stage of its development, that is, prior to the
middle of the third century BCE. Although direct evidence to this
effect is lacking, there exists some circumstantial evidence in the form
of details lacking in one or more of the textual witnesses, which could
have been inserted as glosses in another text.
III. Reconstructed glosses in the textual witnesses of the Hebrew Bible
With real evidence lacking, many glosses are reconstructed on the basis
of circumstantial evidence such as mentioned in the previous section.
Even more instances are reconstructed without any textual basis, and
they are thus in the nature of emendations, although this term is
usually not used in this context. Most of these reconstructed glosses are
little more than a scholarly exercise in ingenuity.
With these clarifications in mind, we now turn to the instances
which are claimed to be reconstructed glosses in the biblical text.
Unfortunately there is no generally accepted terminology or under-
standing for this area. It seems that most scholars use the term ‘gloss’ as
a general term for all elements not extant in the body of the text written
by the original scribe, but added by another hand.22 Some scholars
speak about glosses and ‘other marginal notes’ as two different
categories, while for others the two are identical.*9
22 The definition by Dijkstra, “Glosses,” 55, n. 2, probably reflects the consensus of
scholarship in this regard: “We use a somewhat extended definition of the gloss; not only as
an addition inserted between the lines or in the margin of a manuscript, but also elements
of textual growth inserted in the text base, whether intentionally or unintentionally. As we
will see, it is impractical to make a distinction between glosses proper and expansions in the
text-base because both are found added prima manu and secunda manu.’ Likewise, Klein (see
n. 1), 32 says: ‘A gloss is any kind of explanatory information added to a text by a scribe.’
23 In the older study by Delitzsch, Lese- und Schreibfehler the major category is that of
marginal notes. In his analysis, misplaced marginal notes (‘Dem Schrifttexte einverleibte
Randnoten’) are subdivided into variant readings, glosses (clarifying, correcting, cross-
referring), sundry additions, and scribal/editorial notes. For Delitzsch, glosses are thus a
subdivision of misplaced marginal notes. On the other hand, Fohrer, ’’Ezechiel,” 39
describes the gloss with the minimal designation of a ‘fremdes Einschiebsel,’ and he notes
(without any explanation) that some of the categories of glosses recognized by him referred
to marginal notations and others to interlinear ones.
GLOSSES 63
That in biblical scholarship the use of the term ‘gloss’ is very loose,24
denoting almost anything added to the main body of the text, is also
clear from Driver, ‘‘Glosses.” This study is usually taken as the
standard article on the topic, in which the author constantly and
inconsistently interchanges the terms ‘gloss’ and ‘interpolation’.2° The
background of this confusion is that most of the instances named glosses
in Driver’s study, and in the scholarly literature as a whole, are in fact
reconstructed interpolations.
Some of the reconstructed glosses are genuine glosses as illustrated
(not proven) below, while most of them are no glosses at all.
1. ‘Genuine’ reconstructed glosses are rare
While there are probably very few instances in the biblical text of
what properly may be named (reconstructed) ‘glosses,’ one group stands
out as presenting relevant material, viz., short explanations of names
and words, added to the completed text, either during the textual
transmission, or at an earlier stage.2° For example,
Gen 14:3 .. at the Valley of Siddim—that is (hu?), the Dead
Sea (all textual witnesses).
Gen 36:1 This is the line of Esau—that is (hu?) Edom (all textual
witnesses).
These remarks may have been added in the margin, or directly into the running
text. In the latter case the term ‘gloss’ is used somewhat loosely. Usually textual
evidence is lacking for these glosses, but it exists in the following example,2” in
which the added element was inserted in a wrong place in the text, probably from
24 For similar difficulties with the description of added elements in the Mesopotamian
literature, see n. 8.
The article starts off as following: ’So far as I am aware, there has never yet been any
thorough study of glosses in the Hebrew text.’ But in the next sentences Driver oscillates
between two different terms: ‘Commentators and exegetes have been left to do as they
like, keeping what they will in the text and discarding what they will as interpolations in it
on purely subjective grounds. ... Until a scientific basis, however, has been laid for identi-
fying glosses with a reasonable degree of certainty ... the principles by which glosses may be
recognized ... Two classes of interpolations ... even though these are or may be interpolations
in the widest sense, the text is now often unintelli-gible if they are removed, and they
cannot properly be regarded as glosses.’ (my italics), After these initial considerations the
article speaks only about glosses.
Cf. Driver, ’’Glosses,”124-126 who notes that many such additions are introduced by a
word such as (w®)hu?. Likewise, Fishbane (see n. 1), 44 ff. bases his recognition of glosses on
the employment of what he names (p. 42) ‘technical terms.’
For a similar instance, see Isa 7:17 “wx 75 nN in the context.
64 CHAPTER FOUR
Josh 18:13. OX INV AIAN And bx ANd D171 Own TBI
From there the boundary passed on to Luz, to the flank
of Luz, southward—that is, Bethel (all textual witnes-
ses).
The words ‘that is, Bethel’ refer to Luz, and not to their present place in the
sentence.
Also in the following instance the explanatory note may have been
added secondarily as it lacks in the parallel verse 18:16.
Josh 15:8 MT ... along the southern flank of the Jebusites—that is,
&
Jerusalem (all textual witnesses).
The examples which follow illustrate possible examples of glosses
explaining difficult words (see many more examples collected by
Driver, ‘’Glosses” and Elyoenay [see n. 1]).
Isa 51:17 MT a2y7n7 01D nyap nx (= T V)
... the bowl, the cup of reeling (NJPSV)
ibid., 22 MT -nnn 02D nyap nex (= TV)
... the bowl, the cup of my wrath (NJPSV)
nyap is a rare word, occurring only here in the Bible, and it is not impossible that
it was glossed by the next one being the common word for the same object. There is,
however, no clear versional support for this assumption,78 and furthermore, it is
not clear why this gloss would be written twice in the same context. The two
Hebrew words were probably used as juxtaposed synonyms in their natural way
by the author himself?? and the assumption of a gloss (Delitzsch, Lese- und
Schreibfehler, 136; Driver, ‘“Glosses,” 137; BHS) is therefore unlikely.
Isa 33:21 ON? 0°77 op (all textual witnesses)
... a region of rivers, of broad streams (NJPSV)
28 The word is not lacking in v. 17 in LXX, as claimed by BHS. In that verse LXX contains
two different words for ‘cup,’ though in a construction differing from MT. In v. 22 both
LXX and 5 contain only one word for the two synonymous Hebrew words. However, this
evidence does not necessarily support the claim of BHS for possibly the translators could
not easily find two synonymous Greek or Syriac words or found it not necessary to
juxtapose two similar words in their translation.
Such pairs of synonymous words are often found in the Hebrew Bible (inter alia,
combinations of words in the construct and absolute state, or of two construct words, as
here) and likewise in the Ugaritic literature, as amply shown by Y. Avishur, Stylistic Studies
of Word-Pairs in Biblical and Ancient Semitic Literatures (AOAT 210; Neukirchen-Vluyn 1984).
This particular pair of words occurs also in parallelism in an Ugaritic text (1 Aght 215-216;
see Avishur, ibid., 375).
GLOSSES 65
In this verse it is the first word which supposedly serves as a gloss, explaining the
second one (Delitzsch, Lese- und Schretbfehler, 136; Driver, ‘’Glosses,” 137;
BHS).29
Gen 6:17 PINT 29 O79 21907 NX NIN 717 73N) (all textual witnes-
ses)?!
For My part, I am about to bring the Flood—waters
upon the earth.
BHS designates 0°n (‘waters’), which stands in a loose attributive connection to
the preceding word, as a gloss, with the implication that it has to be removed from
the text.
The same problem exists in similarly phrased verse in the same
context:
Gen 7:6 PINT 2Y O79 77 21INT Aw MND wy jan) (all textual
witnesses)
Noah was six hundred years old when the flood came,
waters upon the earth.
‘Water(s)’ is designated as a gloss by Driver, ‘’Glosses,” 140 and McCarter,
Samuel, 32. The latter uses this verse as an example for explaining the phenomenon
of glosses added in order ‘to explain obscure terms.’94
The preceding analysis has shown that some glosses may have been
added to the text and that vague supporting evidence is available.
Thus, even though no direct manuscript evidence on glosses in Hebrew
sources is known, the assumption of the existence of such glosses at an
early stage of the development of the Hebrew text is not rejected out of
hand. Possibly some convincing examples of this procedure can be found,
but some of the aforementioned cases, which the present writer
considers the strongest ones among the suggested cases, have to be
explained differently.*%
30 However, the two words describe each other, in this case in apposition, and the
assumption of a gloss is therefore unlikely.
1 BHS mentions a Genizah fragment lacking both words (‘flood’, ‘water’).
Whether or not mabbu! should be considered an ‘obscure term’ is hard to determine. It
occurs a dozen times in the Bible, but it is true that the aforementioned two instances are
the first ones to appear in the Bible. ‘Water(s)’ should probably be taken as an apposition to
mabbul. Textual support for the assumption of a gloss is lacking in 6:17 and is unclear in 7:6.
In the latter case BHS and McCarter, Samuel, 33 record LXX as lacking o7n, but in actual
fact this pertains only to manuscript A. The word is found in all other manuscripts, though
in most of them in an inverted sequence (toi USato¢ éyéveto). Rahlfs and Wevers print
the text of papyrus 911 (5 katax\uapde éyéveto bSatoc én THe yc).
33 In the recognition of glosses it should be remembered that the assumed gloss should
always be a word that is more frequent and/or less difficult than the word that is glossed.
66 CHAPTER FOUR
Even less convincing are the following cases for which no manuscript
material is available at all. It is often not clear how the words which
are viewed by Driver as ‘glosses’ explain adjacent or other words in the
context.
Job 30:3 MT AND) ANI WON WP OP rN...
... they flee to a parched land, to the gloom of
desolate wasteland.
Driver, “Glosses,” 131 considers the enhanced pair of words a gloss explaining
mx (’a parched land’). However, that word, which occurs rather frequently in the
Bible, does not seem to be in need of an explanation.
Ps:73:19 MT ON 1D
. Wholly swept away (NJPSV)
According to Driver, ‘‘Glosses,” 137, the second word is a gloss on the first.
However, it is not clear why this word would have to be explained. Furthermore,
the decision which of the two words is a gloss of the other seems to be somewhat
arbitrary. In this case, the recognition of a gloss probably derives from Driver’s
wish to remove one of the two words because of metrical considerations.
Jer 5:15 MT =o N07. O719D CN RT RN
It is an enduring nation, it is an ancient nation.
Without any supporting evidence Delitzsch, Lese- und Schreibfehler, 135,
suggested that the second phrase is a gloss to the first one.
Scores of examples may be added from the analyses of Delitzsch and
Driver of reconstructed glosses which in our mind are no glosses at all.
While the assumption of ancient scribal glosses is not rejected out of
hand, the following should be remembered.
(1) There is no early or late evidence from Hebrew biblical manu-
scripts for the very existence of interlinear or marginal glosses.
(2) The versional evidence adduced in favor of this assumed scribal
habit does not prove the existence of glosses; at most it can be adduced
as direct supporting evidence.
(3) The stylistic pairing of synonymous words such as in the last
examples should be recognized as a literary feature (see Avishur [n. 29])
rather than a basis for assuming glosses.
2. Most reconstructed glosses actually are interpolations
With this negative view on glosses in mind, we now turn to the evidence
on interpolations (exegetical additions). In accordance with the
definitions given above we suggest that most of the examples provided
for (reconstructed) glosses actually are (reconstructed) interpolations.
GLOSSES 67
The major argument for this view is that a gloss, an explanation of a
difficult word, usually provides an explanation meant to remain
outside the syntax of the sentence, while interpolations, exegetical
additions, are integrated into the syntax. We should therefore turn to
the question whether or not there is solid evidence for the existence of
interpolations in ancient sources. Driver, Dijkstra, and Klein (see n. 1)
defined the gloss in such a way as to include interpolations, and by
doing so they provided an answer to the question of the evidence: Since
there is no manuscript evidence for glossing, there is no evidence for
interpolating either. But also if the two phenomena are separated, as
suggested here, Tov, “Textual Base” (see n. 10) does not present any
manuscript evidence for interpolating in the Qumran scrolls. The only
relevant evidence is circumstantial, derived from the ancient versions,
as in the following three examples of interpolations (named glosses):
Gen 14:22 MT ovnw aap yr2y Ox a7 Ox oa onan O70 72 28 OTN N71
PAN}
But Abram said to the king of Sodom, ‘I swear to (the
Lord) God Most High, creator of heaven and earth.’
(= TV)
LXX Sand 1QapGen XXII, 21 lack the word in parenthesis.
SP reads, instead, o°79xN7, ‘God.’
The presumably original text of this verse, reflected in the shorter text of the LXX,
S, and 1QapGen, referred to God as ‘Most High’ (jr2y), a term which also occurs
in Canaanite texts, in which ‘Elyon has the function of ‘creator’ (11?), as here. MT
T V, however, incorporated a single word, YHWH, thus identifying ‘Most High’
with the God of Israel, as if Abram is addressing Him. The original form of the text
is also preserved in MT inv. 19: ‘Blessed be Abram of God Most High, creator of
heaven and earth.’
1 Kgs 8:2 MT All the men of Israel gathered before king Solomon in
Jerusalem in the month of Ethanim (at the Feast—that
is [hu] the seventh month) (= TS V).
LXX lacks the words in parenthesis.
The minus of the LXX, which may also be named a plus of MT TS V, contains the
first mention of ‘the Feast’ (of Tabernacles) in the historical books (cf. Neh 8:14
which mentions that the Israelites dwelt in booths during the feast of the seventh
month). In this verse the LXX differs also in other details from MT.
Josh 2:15 MT(navr xn anna) Amina WP. AN 7D) enn WA aN. ATM
She let them down by a rope through the window (for
her dwelling was at the outer side of the city wall and
she lived in the actual wall).
68 CHAPTER FOUR
The section in parenthesis is lacking in the LXX and its secondary character as an
exegetical addition is assumed by various scholars.4
These examples show that there is some circumstantial evidence for
interpolations in the manuscripts of the Bible. These interpolations
have been integrated into the syntax of the sentence, although in some
cases that syntax limps when the addition entered the text in a wrong
place.
3. Most (reconstructed) interpolations were created at the stage of the
literary development of the biblical books
At this point further clarifications are needed, involving one’s overall
view on the development of the biblical books. The focus of the
discussion has shifted from glosses to interpolations since scholars
usually employed the term gloss for what actually are interpolations.
The employment of the term interpolation is more complicated than
depicted here. Interpolations were not only inserted into texts in the
course of the textual transmission, but similar additions must have been
made at an earlier stage, that of the literary development of the
biblical books. The last mentioned additions should actually not be
called interpolations, but we nevertheless adhere to this terminology
which has been accepted in the scholarly literature since as a rule no
distinction is made between the two levels. We believe that in some
conditions it 1s possible to distinguish between the literary and textual
level with reference to interpolations. It is suggested here that often a
quantitative distinction can be made between the two levels in the
following way. If a number of exegetical additions seem to be connected
with each other in a coherent way, they probably constituted a layer of
additions, created in the course of the literary growth of the biblical
book. On the other hand, if a given addition is not linked with other
ones, if may have been added by a scribe in the course of the textual
transmission, although the alternative solution, that of its creation at
an earlier stage, cannot be rejected. In spite of these doubts, we suggest
that the majority of the exegetical additions recognized by scholars
(usually named glosses) were not created at the textual, but at the
literary level of the development of the biblical books. This pertains to
the examples from Josh 2:15 and 1 Kgs 8:2, mentioned above, as well as
to the examples to be mentioned below. This type of conclusion is based
34 see Tov, “Joshua,”* esp. 334.
GLOSSES 69
on the fact that most additions which are characterized as possible
interpolations somehow are connected with other ones.
In order to explain this point some generalizing statements need to be
made first. The biblical books developed through the expanding of an
earlier form of the composition with new layers, and in the later stages
of this development especially with (a) layer(s) of exegetical
comments. This view has been expressed in modern research, among
other things in the writings of the present author, on Jeremiah, Joshua,
Ezekiel, Samuel, and Kings.2° A good example of this development is
Jeremiah where the short text reflected in 4QJer®4 and the LXX
(‘edition I’) has been expanded with an additional layer (‘edition II’)
in MT.36 Some of these additions are expanded or additional formulae,
others are editorial comments and new material, and again others have
been presented as interpolations or glosses. These instances could have
been presented as interpolations or glosses had they occurred as
individual, scattered phenomena in a different book, but under the
circumstances they should be taken as part of a literary layer. For
example,
Jer 36:6 OMT 71RD (TT 7797 AX 7b) IND WR) 77202 NXP
.. and read aloud from the scroll (that which you
wrote at my dictation, the words of the Lord) to the
people.
Jer 41:1 (7x02) 19n7 on> ow 172K
.. and they ate there together (in Mizpah).
Instances such as these should not be separated from the collective
evidence presented by the textual witnesses in Jeremiah. Although it
would be natural to evaluate each exegetical addition on its own merits
as an individual instance, it would not be correct to do so in Jeremiah in
view of the many pluses in MT. These pluses have to be evaluated as
one unit; some of them may be considered exegetical interpolations, but
most of them form an expanding exegetical layer which stresses certain
aspects of the short text. Some of these are expanded or additional
formulae and personal names, others are editorial remarks,
summarizing statements, and even new data. The analysis thus takes
the quantitative evidence into consideration. In our view, such instances
have often been wrongly isolated from their environment and
evaluated as individual cases of interpolation or glossing. 37
35 Bora summary, see TCHB, chapter 7.
6 Fora summary and other views, see Tov, “Jeremiah.”
For example,
Jer 27:19 muon 9y) or 2D) O° INvA IX MRIY “7 Wx 1D “Dd
70 CHAPTER FOUR
4. Interpolations (glosses) in Ezekiel?
The most frequently discussed book in this regard is not Jeremiah, but
Ezekiel. In this book the deviations from MT in the LXX are less
convincing than in the case of Jeremiah, but its evidence has been more
influential on the analysis of glosses and interpolations in the Bible. In
Jeremiah, the deviations are found in Hebrew as well as Greek sources,
while the data for Ezekiel are culled from the LXX only. In this book,
the deviating Greek translation lacks some 4-5 percent of the words of
MT, and it has been suggested by Tov, ““Ezekiel’’* that the deviating
Greek translation, which is generally very faithful to its parent text,
was not created by the translator who shortened his Vorlage, but
reflected a shorter Hebrew version of the book, which in chapter 7 also
differs recensionally from MT. In previous research, however, the
mentioned data on Ezekiel are closely connected with the discussion of
interpolations in the Bible as a whole (usually named glosses in the
literature).38
It appears to us that the presumed interpolations in Ezekiel should
be taken in their totality as representative of a layer which was added
in the ‘edition’ of MT to the earlier (shorter) edition now reflected in
For thus says the Lord (of hosts concerning the columns, the sea, the
stands) ...
BHS: > LXX, add cf. 52,17
Jer 27:22 BAN “TPS OI nw
(... and there they shal! remain, until the day when | give attention to
them.)
BHS: > LXX, add
Jer 29:16-20 BHS: LXX om 16-20, add; cf. 8a
38 While the discussions of Delitzsch, Lese- und Schreibfehler and Driver, ‘‘Glosses,”
provide many (often identical) examples of interpolations (named glosses) from the entire
biblical literature, the below-mentioned three major discussions are limited to Ezekiel, in
which book the minus elements of LXX are usually viewed as glosses in MT. Of great
influence on the literature was the study by Fohrer, ‘“Ezechiel’”” (1951), followed by those
of Freedy, “Ezekiel” (1970) and Dijkstra, “Glosses” (1986). Earlier studies, likewise on
Ezekiel, less influential on scholarship in general, but of seminal importance for the analysis
by Fohrer and others, had been carried out by Rost (1903-1904) and Herrmann (1908, 1911).
For both, see n. 1. All these studies were limited to Ezekiel, but they referred to a topic
which was to be of general importance for biblical research, that of glosses in the biblical
text. It was surmised, probably unconsciously, that a similarly large number of glosses such
as found in Ezekiel (364 according to Fohrer) must have been contained in other books as
well. Ezekiel, however, probably presents a special situation, and thus if the view about the
many glosses in that book appears to be ill-founded, the views about other books need to be
adjusted as well.
GLOSSES 71
the LXX.3? With Dijkstra we disregard the many reconstructed
interpolations in Ezekiel, most of which are insufficiently founded. The
evidence of the LXX is taken as representing a different Hebrew text
(edition), since the translation is always faithful. This Hebrew
Vorlage of the LXX is almost always shorter than MT. In other words,
MT often reflects late exegetical elements, but these elements should
not be viewed as individual elements inserted into the text, that is, as
interpolations or glosses, but as representatives of an extensive layer of
exegetical additions.
Like in the case of Jeremiah, some of the plus elements of MT in
Ezekiel could represent individual interpolations (others: glosses) and
they have indeed often been presented as such (see n. 38). The examples
which follow were named ‘explicative glosses’ by Freedy, ‘’Ezekiel,”
137-141 (the first example), and ‘exegetical glosses’ (the last two
examples [146-149]). All elements lacking in the LXX are indicated by
italics and parenthesis.
Ezek 8:3 MT (nen-ii7) Www nd NX
to the entrance of the (inner) gate?
Ezek 1:22. MI (NTA) NAPA PVD
with an (awe-inspiring) gleam as of crystal
Ezek 3:18 MT (Aywrn) AID yw WAI?
to warn the wicked man of his (wicked) course
The plus of MT in this verse as well as in the next one are probably prompted by
the implication of the passage as a whole.
Other plus elements of MT in Ezekiel, however, cannot be considered
interpolations, viz. parallel words (the first three examples) and new
material (the last two examples).
Ezek 5:14 MT Jrpma°a0 Wwe oa (757N4)) 7AIN? Jinx1 (= TS V)
I will turn you into a ruin (and a reproach)
among the nations that are around you.*!
39 In his description of the glosses, Fohrer does not distinguish between glosses which
are supported by textual evidence (LXX) and glosses which are not supported, a
distinction which was to be made later by Freedy and Dijkstra.
40 The addition in MT is wrong in the present context. ‘It is premature in terms of the
itinerary of the divine tour as represented by this stage in the narrative’ (Freedy, “Ezekiel,”
138).
41 For a similar situation, see Jer 49:13 (29nd) aD7Nn> nw *D, where the word in
parenthesis is lacking in the LXX.
72 CHAPTER FOUR
Ezek 5:15 MT man nindn3) (79nI1 4X3) ayy 72 °*mwya (= TS V)
when I execute judgments against you (in anger
and in fury) and by chastisements of fury.4?
Ezek 6:6 MT (inawi) Mawn osninam...(=TS V)
... your altars, and shall be ruined (and shall
be desolate)
Ezek 16:13 MT (FD2N9 -n2EN)) IND TNNI °D"M (= TS V)
You were very very beautiful (fit to be a queen).
Ezek 20:28 MT Omnia ns ow nari nay py 2a) ANT AVI 4D INT)
(O32? OVD Ow UN) (= TS V)
They saw every hill and every leafy tree, and
there they made their sacrifices. (There they
placed their vexatious offerings.)
See further 1:22 quoted above.
Because of these five examples and many additional ones (for a full
discussion, see Tov, “Ezekiel’*), the presumed interpolations in Ezekiel
(named glosses in the literature) should be taken in their totality as
representative of a literary layer, added in the ‘edition’ of MT to the
shorter and earlier edition of the LXX. Accordingly MT might indeed
reflect exegetical additions; however, these additions should not be
viewed as individual elements, but as components of a separate layer.
Examples like the ones adduced here thus do not prove that the book of
Ezekiel abounds with many interpolations or glosses. For one thing, it
would be quite unnatural to assume that this book was interpolated to
such a great extent (see n. 38).
5. Other books
Many of the differences between MT and the LXX in Joshua are
explained in a way similar to those in Jeremiah in Ezekiel. The Greek
text of that book provides now a shorter and then a longer text, often in
details which have been recognized as significant for the literary
history of the book. Thus the aforementioned longer text of MT in Josh
2:15 does not bear the character of an individual exegetical remark,
subject to textual evaluation, but it is part of a literary stratum. This
pertains also to the following example, in which the minus of the LXX
is enclosed within parenthesis.
Josh 1:15 MT nwa od? yn) Wwe (AMX ONw7) OoNw7? pax? onIWw?
(7 Tay)
42 Cf. Deut 29:27 where ann XI occur in a similar context.
GLOSSES 73
Then you may return to your inherited land (and you
shall take possession of if) which Moses (the servant of
the Lord) gave unto you.
In this verse, listed by Weingreen (see n. 1) 151 as an example of a gloss, the first
minus of the LXX, serving as a plus in MT, is contextually an addition in the text,
disturbing its syntax. However, the plus, which does not improve the meaning of
MT, and hardly provides elements that are exegetically significant, cannot be
considered a gloss, although with some difficulty it could be viewed as an
interpolation. As an interpolation, however, it is not the individual case which is
of interest, but the fact that in Joshua, MT (as opposed to the LXX) contains many
more pluses of this type,*¥ all of which together form an added layer.
IV. Conclusions
The course which our investigation has taken is quite unusual. The
discussion of the categories of scribal additions in ancient sources as
well as of individual glosses and interpolations in the manuscripts of
the various biblical books has led to negative conclusions with regard
to the existence of these phenomena in the manuscripts of the Hebrew
Bible. It was realized that the scholarly discussion of individual
instances of glossing has been influenced much by a vast body of minuses
in the LXX of Ezekiel and secondarily also of Jeremiah, which have
often been described as glosses in the scholarly literature. However,
the textual data which look like scribal glosses or interpolations in the
MT of these books actually were probably created at an early stage of
the history of the biblical books, and not as individual phenomena. In
other words, the data in Ezekiel and Jeremiah which have been so
influential on the research on glosses and interpolations elsewhere in
the Bible, probably are irrelevant to the discussion of scribal glosses
and therefore misleading. The final conclusion on the existence of
glosses and interpolations in biblical manuscripts must therefore be
quite negative. More in detail:
1. No glosses proper (as defined in n. 2) have been detected in the
manuscripts of the Hebrew Bible. Evidence for assumed glosses in the
tradition of the Hebrew Bible, deriving mainly from the versions, is
circumstantial and uncertain.
43 Cf. Tov, “Joshua”*; H.M. Orlinsky, ““The Hebrew Vorlage of the Septuagint of the
Book of Joshua,” VTSup 17 (1969) 187-195.
74 CHAPTER FOUR
2. The many interlinear and marginal additions in the Qumran
scrolls are mainly in the nature of corrections and not of glosses or
interpolations.
3. Almost all glosses suggested by biblical scholars are reconstructed,
while some are supported by evidence in the ancient versions.
4. Most elements which have hitherto been described as glosses,
actually are interpolations (exegetical additions).
5. While some of the elements recognized as interpolations must
have been added at the scribal level, many (if not most) of them
belonged to a layer (edition) added to an earlier text (edition) at the
stage of the literary growth of the biblical books.
6. The many plus elements of the MT in Ezekiel (as compared with
the LXX) are no individual glosses or interpolations, as claimed by
several scholars, but a layer of additions in this book added to an
earlier edition.
7. The fact that many elements in the MT of Ezekiel (as compared
with the LXX) have been described as glosses in several studies from
1903 onwards has influenced scholars to believe that biblical
manuscripts contained a great number of glosses (or interpolations).
CHAPTER FIVE
EINE TORA FUR DEN KONIG TALMAI
G. Veltri’s study, Eine Tora fiir den Kénig Talmai! deals with the
evaluation of and references to biblical translations, more especially the
LXX, in rabbinic and Jewish-Hellenistic sources. This detailed study
provides not only the rabbinic and Jewish-Hellenistic sources
themselves, but also extensive analyses of these sources and of the
Forschungsgeschichte on them. This book seeks to solve mainly one
question which is posed on p. 18, and the answer to which is provided in
the summary on pp. 215-219, although the author’s views are also
expressed throughout the book (e.g., pp. 107-112).
As explained in the introduction (pp. 1-18), the author wishes to
show that all scholars before him have wrongly embraced the view that
the Jews rejected the LXX, either because the Christians had adopted that
version or because they were opposed to the Hellenization of other
cultures by King Ptolemy who had the Hebrew Bible translated into
Greek for precisely this purpose. That Veltri’s presentation of the views
of some of his predecessors is imprecise 1s shown below.
Veltri’s study attempts to disprove this view, which according to him
has become the accepted view of the scholarly world. He realizes (p. 19)
that the rabbinic sources actually do not contain explicit information on
the acceptance or rejection of the LXX, and that they only contain some
traditions from which such information may be culled. Veltri’s main
contention is that when these traditions are properly analyzed, they do
not prove the rejection of the LXX by the rabbinic sources.
The sources which are analyzed at length are (1) a number of
traditions about verses ‘written’ or ‘altered’ ‘for king Ptolemy’; (2) a story
about the creation of the Greek translation of the Torah, in different
versions, sometimes in connection with a dictum that the Torah cannot
be translated at all, in any language (except for Greek). In some traditions
elements (1) and (2) are juxtaposed. According to Veltri these traditions
1G Veltri, Eine Tora fiir den Konig Talmai—Untersuchungen zum Ubersetzungsverstandnis in
der jlidisch-hellenistischen und rabbinischen Literatur (TSAJ 41; Tubingen 1994).
76 CHAPTER FIVE
do not explicitly refer to the acceptance or rejection of the LXX by the
rabbis, although they do provide background information on the
rabbinic understanding of the translation procedure of the LXX.
The LXX was basically a Jewish translation produced by Palestinian
Jews or Egyptian Jews with a strong Palestinian background. This issue
may be irrelevant, since at a certain moment the background of this
translation was apparently forgotten when the LXX came into disuse. But
it should have been stressed by Veltri that the LXX was a product of
Egypt, used by Alexandrian Jews, who were not fluent enough in
Hebrew to use the Hebrew Bible. They had no alternative, so to speak,
but to use a translation, while the Jews of Palestine and Babylon did have
an alternative, and could allow themselves to disregard the Greek
translation. Veltri does not discuss these matters, nor does he address
any internal evidence regarding the use or lack of use of the LXX in
Jewish writings of the Second Temple period, including the rabbinic
writings, in the latter case, the lack of use. Veltri merely studies the
approach to the LXX from external evidence, viz., a small list of
references to verses in the LXX and a story about the creation of the
translation. This self-imposed limitation determines the nature of the
conclusions reached which, albeit important, are limited to the issue of
how the LXX (and the Targumim) are evaluated in the rabbinic writings.
The conclusions are important for our understanding of rabbinic Judaism
and for the way traditions are created and transmitted within the
rabbinic literature. However, they do not necessarily apply to the Jews of
Palestine or Babylonia as a whole. Thus the author does not discuss the
discovery of early fragments of the LXX (and its early revisions) in
Qumran.
But more is involved. The self-imposed limitations of Veltri lead him
to turn to only one of the exponents of rabbinic Judaism, viz., rabbinic
literature. He rightly suggests that we should not be quick in assuming
on the basis of that literature that the rabbis rejected the LXX. But there is
another type of evidence which also pertains to rabbinic Judaism, and
which leads us to believe that certain rabbinic circles did reject the LXX.
This evidence is provided by an internal analysis of the early, pre-
Christian revisions of the LXX. These revisions, revising the LXX in
accordance with an ever-changing, proto-Masoretic (proto-rabbinic) text,
reflected the need to use a Jewish-Greek text based on the content of the
Hebrew Bible, often different from that of the Greek Bible. Several of
these revisions antedated Christianity, so that Christian influence could
not have been instrumental in their creation (kaige-Th [reflected among
other things in 8HevXIlgr], 7QLXXExod, P. Oxy. 1007, and P. Rylands
EINE TORAH FUR DEN KONIG TALMAI 77
Gk. 458). This revisional activity shows that certain Jewish circles moved
away from the LXX in the pre-Christian era, as mentioned by the present
reviewer in TCU, 143 (and not as quoted by Veltri, p. 18). Whether or not
these circles were identical with the ones from which the rabbinic
traditions derived is not known, but it is likely that they were closely
related. Note, for example, that kaige-Th’s exegesis is described by
Barthélemy in the subtitle of his Devanciers as ‘sous |’influence du
rabbinat palestinien.’ This type of evidence should have been discussed
by Veltri, who would probably have come to the conclusion that the
issues are more complicated than presented by him. Thus, the (later)
Jewish translation of Aquila (also a revision of the LXX) is often quoted
in the Talmud (Veltri, pp. 186-90 and Reider, quoted on p. 186), while the
individual renderings of the LXX are not. This evidence alone leads us to
believe that the rabbis were not content with the LXX, to say the least.
In the introductory pages, Veltri points out the difficulties inherent in
the Epistle of Aristeas and in the stories in the rabbinic literature.
According to both sources, the LXX was made on behalf of the
Ptolemies—indeed the story that an earlier AXayue¢ in the LXX was
changed in Lev 11:6(5) to Sacvmouc is understandable only when that
change was made within the framework of a translation for the
Ptolemies, who the translators allegedly did not want to insult by
including Aaywdic among the unclean animals. Thus, while rabbinic
traditions recognized the Ptolemaic origin of the translation, at some
stage that translation was rejected. But Veltri’s new insights lead us to
believe that this rejection took place only at a late stage in the rabbinic
tradition and that at first the LXX was embraced by the rabbis as a
legitimate source. Veltri shows that the often-quoted dictum comparing
the creation of the LXX with the making of the golden calf appeared in
the late tractate Sopherim (1.7) and not in earlier sources. Likewise, the
day of fast in memory of the preparation of the LXX translation is not
mentioned before the late Massekhet Ta‘anit.
Veltri attempts to correct the imprecisions in the recording of the
rabbinic evaluation of the LXX and it seems to us that Veltri has
succeeded in doing so through his analysis of the stories about the
creation of the LXX in rabbinic literature. There is indeed no explicit
reference in the early rabbinic literature to the rejection of the LXX. But
the evidence remains open to different interpretations, as we shall see
below.
Veltri sets out to prove his thesis by two main arguments.
1. One of the arguments which has been used in the past for proving
that the rabbis rejected the LXX derives from an interpretation of lists of
78 CHAPTER FIVE
details in which the rabbinic text of the Bible (the so-called proto-
Masoretic or proto-rabbinic text) differs from the LXX. These lists have
always been viewed as lists of alterations (note that 117-w and ow are
used in some rabbinic passages, as opposed to 12n> and 0°737 in others).
Since the story of the creation of the LXX is mentioned in b. Meg. 9a in
conjunction with the list of these passages (1? 12n1), it had been
suggested before Veltri that the rabbis resented these differences, that
they considered them ‘alterations’ by the Greek translators, and that they
therefore rejected the LXX. However, they were not ‘alterations’ (a7 1-w),
argues Veltri, but merely passages (0°727), and only in the later tradition
they were considered alterations. Since these readings were originally
not alterations, according to Veltri they could not be taken as an.
argument in favor of the view that rabbinic Judaism at an early stage
rejected the LXX.
The passages in the early and late rabbinic literature are presented
with all the manuscript variations on pp. 220-247 and are analyzed in
detail on pp. 22-109. A great part of the book is thus devoted to a
detailed analysis of these passages. The passages have been transmitted
either separately or in lists containing between ten and eighteen such
items. Major problems regarding the nature of these passages (their
number, original language, agreement with the LXX) had been discussed
by scholars preceding Veltri One major obstacle for a coherent analysis of
these items is that most of the supposed readings of the LXX are not
known from any of the manuscripts of that translation even though
according to the rabbinic sources they were included in the LXX. This
reviewer has therefore suggested that the original Greek readings have
been lost, and that they existed once in an earlier (original?) version of
the LXX (Tov, “Alterations”*). Veltri himself returned to a view which
had been suggested previously in various forms (Frankel, Friedmann,
Aptowitzer, Talmon) that the changes actually do not refer to differences
between Hebrew and Greek texts, but that they pertain to inner-Hebrew
variations. More specifically, according to Veltri most of the ‘readings’
which according to rabbinic tradition were written on behalf of King
Ptolemy actually reflected rabbinic exegesis of some kind. In a lengthy
discussion (pp. 22-109), Veltri points to such exegesis in rabbinic sources,
not necessarily identical to the list of readings/alterations, but at least
referring to the same biblical verses. At the same time, Veltri realizes that
four instances which have been transmitted in ‘some manuscripts of
some rabbinic tractates’ (p. 98) must be regarded as translations from
Greek into Hebrew (Gen 2:2; Exod 12:40; Lev 11:6; Num 16:15). The other
ones reflect ‘merely text-critical and exegetical difficulties in an already
EINE TORAH FUR DEN KONIG TALMAI 79
fixed Hebrew text’ (p. 98). To some of these difficulties Veltri finds
references in different rabbinic sources, explicit or implicit, while for
other ones he does not find such references. For example, the different
sequence of the text written ‘for King Ptolemy’ in Gen 1:1 (‘God created
in the beginning’) reflects problems raised by and solutions given in
various rabbinic commentaries (Gen. Rab. 1:14; Tanh. Buber Bereshit 4)—
see pp. 25-31. The addition in Deut 17:3 ‘for King Ptolemy’, 071”, is
paralleled by an identical addition in Siphre Deut. 148 (see pp. 92-7).
Likewise the addition of 17x77? in Deut 4:19 ‘for King Ptolemy’, is
paralleled by an identical addition in the late midrashic collection
Warethanan ad loc. (pp. 92-7). The change ‘for King Ptolemy’ of 7 to 011
in Gen 49:6 reflects an inner-Hebrew development, identifying Tw as 1Ww
(not Ww) and applying it to Joseph; possibly 012 is an orthographical
variation of 0°)X (Apis) or 0°)x 10 (Serapis), identified with Joseph (b.
‘Abod. Zar. 43a). See pp. 63-69.
But the principle, rather than the details, are important in this
analysis. It remains difficult, and actually unexplained, how and why
difficulties in a biblical verse which one or more rabbis present according
to some source should be ascribed to the translational activity of the
seventy translators.
More importantly, whether or not the very difficult problem of the
original language of the changes ‘for King Ptolemy’ can be solved may
not be relevant for the main thesis of this book, namely Veltri’s view that
the rabbis did not reject the LXX. Even if the changes/readings ‘for King
Ptolemy’ had originally been phrased in Greek, Veltri could probably, on
the basis of the arguments mentioned below, still maintain his main
thesis that the LXX was not rejected. The argument would be stronger if
the changes/readings ‘for King Ptolemy’ were not based on Greek
readings, as Veltri suggests, but it makes no major difference. The
suggestion that the readings/changes are inner-Hebrew did not convince
the reviewer, in spite of the parallel with the development of the lists of
tiqquné sopherim also containing inner-Hebrew changes, mentioned on p.
105, n. 326, but not further developed (and in spite of the parallel of the
Qeré notations in the Masorah, not mentioned by Veltri). In both of these
cases, lists of phenomena of a different origin were combined under one
heading. Thus only some of the tiqquné sopherim listed in the rabbinic
literature are corrections of the sopherim, while others are mere exegetical
euphemisms. Likewise only some of the Qeré notations originally had an
authoritative status as corrections, while others originally were probably
mere variae lectionis subsequently upgraded to the status of a Qeré. By the
same reasoning one could argue that only some of the changes ‘for King
80 CHAPTER FIVE
Ptolemy’ were real Greek renderings, while the other ones, actually cases
of inner-Hebrew exegetical changes, had nothing to do with the LXX.
This is a possibility, which seems to me remote, but the option of this or
that explanation does not affect the main thesis of V with regard to the
rejection of the LXX.
The main thesis of Veltri, described on pp. 107-12, relating to the lists
of readings/changes of the LXX, is that these were originally indepen-
dent readings, sometimes combined into clusters of two, three instances,
and only later joined (by the sopherim) to the lists which are now found in
several places in the rabbinic literature. The background of these
readings/changes is that they were actually written ‘for King Ptolemy,’
the one on whose behalf the exegetical changes were inserted in the
translation. This is a very central point in the argumentation of Veltri,
from which the book derives its name: Eine Tora fur den Konig Talmat.
That is, the rabbis prepared a written midrash for King Ptolemy since he
did not have the advantage of studying Torah with the rabbis (p. 108).
For the rabbis this written Torah was the LXX! That the LXX contained
such an exegetical copy of the Torah can also be inferred from the use of
the term 127, introducing the individual readings/changes (ja 4nX 71
Pan wn2n? ianaw/irwy 072777), parallel to the term 7nx 727 introducing
an alternative explanation in rabbinic literature. According to Veltri, the
original tradition spoke about ‘writing’ to Ptolemy, secondarily altered to
‘changing’ (p. 108).
Although according to Veltri, some of the rabbis knew Greek, and
some of the readings/changes derived directly from the LXX, the rabbis
conceived of them as reflecting a different form of exegesis, so that the
possibility of a different Hebrew Vorlage of the LXX never entered their
minds. Veltri does not explain how the readings/changes, which
Originated in the exegetical tradition of midrashic possibilities in
conjunction with the Hebrew Bible, came to be ascribed to the Greek
translation (cf. Tov, “Midrash Joshua”*).
2. On the basis of the lists and stories about the creation of the LXX,
chapter 2 established that the changes/readings reflect midrashic
possibilities and that the LXX is a translation made for King Ptolemy
personally. The next chapter (3) goes one step further and investigates
the approach of the rabbis towards translating and exegeting in general,
also beyond the Greek language. For scholarship this is a novel
discussion in which Veltri makes some very important observations and
distinctions not recognized previously, using only external, and not
internal, evidence, as in chapter 2.
EINE TORAH FUR DEN KONIG TALMAI 81
Thus Veltri noticed that the verb used for the activity of the LXX
translators is 1nD as opposed to that used for the Aramaic translations,
viz., 0.0n. The use of 1n2, to write, implies that for the rabbis the LXX did
not constitute a regular translation from the source language to the target
language, but the writing in a Greek shape of the content of the Hebrew
Bible. The Greek and Hebrew versions were considered to be of equal
value—just like the understanding of the LXX in several Jewish-
Hellenistic sources. The fact that the LXX was made for King Ptolemy,
and not for the liturgical needs of the Jewish community, is stressed time
and again in the rabbinic sources, implying that the exegetical changes
were meant to make the king’s reading easier. This aspect of the rabbinic
tradition, which hitherto was taken cum grano salis, is taken seriously by
Veltri who provided the background to these statements.
In this regard Veltri recognizes a major distinction between the
rabbinic approach to the LXX and to the Aramaic Targumim. In the
rabbinic sources the former translation was meant for external use, for
the King, while the Targumim represent an internal product of Judaism,
produced and controlled by the rabbis. For the latter, the verb o17Nn is
used, not only with regard to the Targumim (see p. 181), but also with
regard to other types of translations. Although we know the Targum to
be a special type of translation, for the rabbis 0)7n referred to the act of
translating in general, while 2n> (with reference to the LXX) was used for
the writing of the Bible in Greek. In accordance with this usage, the
seventy translators are not named translators in rabbinic sources (rather
‘old men,’ ‘wise men’), and only in the later Christian tradition were they
called ‘translators’ (p. 193). On the other hand, the meturgeman was a
translator, or actually a mediator between the official exegetes of the
Bible (the wise men or the rabbis) and the community.
The only other translation for which o)7n is used in rabbinic sources
is that of Aquila. Veltri ascribes this situation to the fact that Aquila, like
the Targumim, reflects rabbinic exegesis (p. 212). This conclusion,
however, may be hasty, since the mentioning of Aquila cannot be
contrasted with other translations since they are not mentioned in the
rabbinic literature. Furthermore, the case of Aquila is unique because of
the confusion between 0?°"7y and 0197nx in rabbinic literature. For the
latter the formula 01927)x OWN or OF7PX O17°N was in order.
Another difficulty with Veltri’s stringent distinction between 1n> (for
the LXX) and onn (for the Aramaic Targumim and Aquila) is that in the
later tradition the distinction became blurred when o17n came to be used
also for the LXX (Sopherim 1.7 and Sepher Torah 1.6).
82 CHAPTER FIVE
The author’s Riickblick (pp. 213-214) and conclusion (215-219) stress that
his main aim was to disprove the idea of the rabbinic rejection of the
LXX. However, the material analyzed in this book pertains to many more
issues than just the question of whether or not the rabbis rejected the
LXX. If Veltri had wanted to address only this question which he
considers the main focus of this book, he should have written a more
limited monograph. But Veltri considers his task to be much more
comprehensive, so that he helps us to understand many more issues. At
the same time, the structure of the book would have been improved if
Veltri had posed his questions differently. For even if one disagrees fully
or partially with Veltri’s main thesis, one could still benefit much from
other aspects of this very valuable book.
Veltri demonstrated convincingly that in the early rabbinic sources
the LXX is presented differently from what has been thought hitherto. In
these sources the LXX is indeed presented as ‘eine Tora fur den Konig
Talmat’ (thus Veltri; or rather ‘die Tora’?). However, it 1s not clear to
what extent the information contained in the list of readings/changes
can be used to describe the nature of that Torah as a personal exegetical
copy intended to help Ptolemy in understanding the Torah since he was
deprived of rabbinic exegesis. The early rabbinic tradition as embedded
in rabbinic sources probably did not reject the LXX. This has been
established by Veltri, but at the same time there is also other evidence
(relating to pre-Christian revisions) showing that certain rabbinic circles
discontinued the use of the LXX (see above). The picture is thus more
complicated than suggested by Veltri. That later rabbinic sources
(Sopherim, Sepher Torah, Massekhet Ta‘anit) present a negative evaluation
of the LXX is known to Veltri and on the last page of the book before the
conclusion (p. 214) he states that he does not know how these sources
reached such a negative view. It seems to me that the earlier traditions
about readings/changes for King Ptolemy contained that criticism in a
seminal way, which could have influenced the later tradition.
Furthermore the embracing of the LXX by Christianity (pace Veltri, p.
215) and the replacement of the LXX by Jewish revisions in Jewish circles
(BCE and CE) provided all the elements for such a negative view.
H. LEXICOGRAPHY
CHAPTER SIX
THREE DIMENSIONS OF WORDS IN THE SEPTUAGINT
1. The LXX and post-Septuagintal literature
The LXX is a Greek text, and accordingly, its words should be
investigated within the framework of the Greek language. However,
the latter procedure alone cannot be satisfactory due to the inclusion
within the language of the LXX of many non-Greek elements derived
from the source languages. These elements must therefore be analyzed
in the light of that translation and, as a consequence, the LXX deserves
special attention within the lexical analysis of the Greek language.
Lexicographers analyze words in languages and literatures with the
aim of describing their meanings in a dictionary. This task is not easy
with regard to ancient literatures where there are no native speakers to
be consulted. Equally difficult is the lexicographical description of
translated words, as the language of a translation is often unnatural.
These two difficulties are combined in the lexicographical! description
of an ancient translation—in our case, the LXX.
The issue under review is how and at what level are meanings of
words in the LXX determined. Meanings of words in literary
compositions are ascertained on the basis of both linguistic and
contextual data. It is probably true to say that if an author wanted a
word X to be understood by the readers as meaning a, then the meaning
of that word X within the context under consideration is a. Such
meanings can often be established by an analysis of the author’s
intentions. By the same token, words in a translation should be taken in
the way in which they were intended by the translator. Thus, in very
abstract terms, the lexicography of a translation aims at recovering the
meanings of the words that were intended by the translator(s). This
definition will aid in the deciding of several practical issues, such as
the one following.
A tension can often be recognized between meanings of words intended
by a Greek translator and meanings attributed to the same words in the
writings of the Church Fathers and in the translations made of the
LXX. Three examples follow.
86 CHAPTER SIX
Gen 1:16 MT) 92790 nownn? TopA TNT NI OTT N2WNN? DTA RT NR
LXX TOV dwoTpa TOV Léyav ele dpyde Tic hwépac Kal Tov
gwotipa Tov éddoow ele dpyde tie vuKtécg
La (et fecit deus dua luminaria maiora) luminare maius in
initium diei et luminare minus in initium noctis!
dpxt\ is used in the LXX in many senses, especially since wx4, its main
equivalent in the source language, occurs in the Bible with a variety of
meanings. As a result, several occurrences of dpyt in the LXX can be
understood in different ways. For example, although dpyxt in Gen 1:16
was undoubtedly meant by the translator as ‘governing,’ ‘regulating’—
cf. its Hebrew counterpart n>wnn>*—, the context also allows for other
explanations. Thus the Old Latin translation (La) took this dpxt as
‘beginning’ in accordance with its most frequent use in the LXX, a
meaning which occurred also earlier in the chapter (Gen 1:1). Although
the rendering initium in Gen 1:16 is understandable within its context, it
does not represent the Greek translator’s intention.?
Jer 17:27 MT OPwWIT MYON WIN}
LXX Kal kataddayetat dudoda | epovoaAnu
Lapud Tyconius4 = et consumet itinera Hierusalem
dudodov is used in Greek both as a ‘block of houses surrounded by
streets,’ and more frequently as ‘street,’ the latter especially in papyri,
although apparently not before the Roman period. The former meaning
was intended by the translator of Jeremiah, for dudoSov, when taken
thus, appropriately represents ]107x.° La, however, took the Greek
word as ‘street,’ a sense which was contextually plausible.®
Ps 23(24):10 (and elsewhere in the book):
MT MIE WW?
LXX KUpPLOG TUV Suvd Lew
Ps. Rom., Ps. Gall. Dominus virtutum
The standard equivalent in the LXX of N23, i.a. when used as ‘army,’ is
S5uvaptg (see LSJ, s.v. I. 3 for parallels in secular Greek). 50vapte is also
used in this meaning in the phrase ktptoc tav Svvdyewv which renders
1 See B. Fischer, Vetus Latina, vol. 2, Genesis (Freiburg 1951).
2 The plural nouns in the LXX probably reflect Mwnn?.
3 See S, Lundstrom, Ubersetzungstechnische Untersuchungen auf dem Gebiete der christlichen
Latinitat (LUA NF I 51, 3; Lund 1955) 116-128 for additional examples of a misunder-
standing of apx7 in La.
4 See F.C. Burkitt, The Book of the Rules of Tyconius (TS I, 1; Cambridge 1894) 62.
5 This word appears to have been difficult for the translators, see Tov, “Understand.”
6 EA. Sophocles, Greek Lexicon of the Roman and Byzantine Periods (New York 1900), s.v.,
similarly explains the meaning of dd¢o8ov in Jeremiah as ‘street.’
THREE DIMENSIONS 87
NIXIN 717° on 16 occasions. In Ps. Rom. and Ps. Gall., however, the Greek
phrase has been misrepresented by Dominus virtutum, in accordance
with the more frequent meaning of Stvapic.’ This understanding can be
contrasted to that of Jerome: Dominus exercituum.
The common denominator of the three above-mentioned examples is
that the words under review are polysemous—.e., they were used in
different senses in Greek—and that they, almost by implication, were
also interpreted in different ways. Amidst the plurality of internally
possible interpretations of a given word in the LXX, the only correct
interpretation is often indicated by the Hebrew source.
The examples establish beyond doubt the existence of at least two
different dimensions of biblical words, viz. the meaning of a biblical
word as intended by the translator in a given context, and the different
meanings which were applied to that word after the completion of the
translation.® To be sure, this distinction is made with regard to all texts
which have been interpreted.
LXX lexicology must concentrate on the intentions of the translators,
mainly by an analysis of the translation techniques employed.
However difficult it may be to grasp these intentions, only they
determine the real meaning of words in the LXX. Consequently, while it
is interesting to study meanings which were applied to biblical words
by later generations, and, while such information is often helpful for
establishing the meanings of the biblical words themselves, by its very
nature this is a secondary source for LXX lexicology.
A distinction was thus made between meanings of words intended by
the translators and meanings attached to the same words after the
completion of the translation. However, even within the LXX such
developments may be detected. In some cases one can distinguish
between two dimensions of meanings which were intended by the
translators in different contexts (in addition to a third dimension after
the translation).
2. Lexicography and translation technique
An analysis of lexical Hebraisms should help us in determining the
nature of the lexicographical description. A Hebraism? may be defined
7 Cf. M. Flashar, “Exegetische Studien zum Septuagintapsalter,” ZAW 32 (1912) 81-116,
esp. 90.
Meanings of this type may be recognized in all sources which regard the LXX as a
Greek text, in isolation from its Hebrew source, such as the Church Fathers and the
daughter versions of the LXX.
? The bibliography on this aspect of the language of the LXX is very extensive. Early
studies are discussed by J. Ros, De studie van het Bijbelgrieksch van Hugo Grotius tot Adolf
88 CHAPTER SIX
as a Greek word, phrase, or syntagma which expresses certain charac-
teristic Hebrew elements in Greek in an non-Greek fashion. Sometimes
an isolated parallel to the Hebraism may be spotted in a secular Greek
source, but the word or element should nevertheless be considered a
Hebraism if the great frequency of its occurrences shows that its
appearance is conditioned by Hebrew rather than Greek usage. This
phenomenon is closely related to the translators’ approach to the
technique of translating, that is, the occurrence of a Hebraism is a
direct result of the system of stereotyped (automatic) representation of
Hebrew words in the LXX, For a detailed description, see TCU, 20-23.
Since the consistent representation of Hebrew words by one Greek
equivalent was often more important to the translators than context-
ually plausible renderings, their technique was bound to do injustice to
several Greek words. For the translators also often used a stereotyped
equivalent when the meaning of the Hebrew did not suit that of the
Greek. In this way non-Greek elements, usually named Hebraisms, were
introduced into the vocabulary of the LXX.
At the level of lexicography, Hebraisms do not function as ordinary
Greek words possessing Greek meanings,!9 but they are used as mere
symbols representing Hebrew words, as in the case of o1>w and elpfvn.!!
Not infrequently 07>w is used not only as ‘peace,’ but also as ‘welfare’
and ‘health,’ and these meanings should have been rendered into Greek
by words other than elpt\vn. Nevertheless, the wish for stereotyped
representation often led a translator to render such occurrences of 01?v
also with elpyvn. E.g., 2 Sam 11:7 aan?nn ow... Ow) - ele elpyunv
Tov ToAépou and Judg 18:15b ow) 99 ~oxwr - kal Aputyoav abtov elec
elpynv.!2 Would it be correct to record in a Greek dictionary ‘welfare’
or ‘health’ as a special meaning of this elprjvn, explaining it as ‘to greet
a person, inquire after their health’ (thus LSJ), on the basis of evidence
from the LXX only?
Deissmann (Nijmegen 1940). Later studies are analysed by Lee, Lexical Study, ch. II. See
further U. Rapallo, Calcht ebratci nelle antiche versioni del “Levitico” (Roma 1971); Walters, Text,
143-154; Tov, TCU, 22-24.
For this terminology, cf. the title of D. Hill, Greek Words and Hebrew Meanings
(Cambridge 1967). See Tov, “Greek Words.”*
1] In the LXX, 077 is rendered by elprvn in 178 instances and further by 18 different
equivalents all of which occur only once or twice. Conversely, elpvn represents nearly
exclusively o1>w. elprvn is thus the main equivalent of n1>w, often used automatically. The
choice of elprywn was natural since the most frequent meanings of 07°w, ‘peace,’ “peace from
war,’ and ‘tranquility’ are well represented by the most frequent meaning of the Greek
word, viz., ‘peace from war.’
2 Contrast the contextual translation equivalent of LXxA- Kal Hotrdoavto avtov.
THREE DIMENSIONS 89
Invoking the principle that LXX lexicography must endeavor to record
the meanings which were intended by the translator(s), we suggest that
some translators did not use elpyvn in accordance with ordinary Greek
usage, and that they did not have a definable meaning of elpfvn in
mind. They simply equated 01?wv with elprivn on a practical level.
Consequently, one might say that for many of the translators elpyvn
was merely a symbol representing 01>w. Accordingly, a Greek lexicon
should not create anew meaning of elptyvn on the basis of the LXX alone,
claiming, as it were, that the translators enlarged the semantic range
of etpr\vn. Such a claim cannot be made, for most translators were
probably not aware of the semantic implications of stereotyping.
Two notes are appended to this analysis.
1. Beyond the LXX the Hebraistic use of elptivn and of similar words
occurs also in the New Testament and in other sources which were based
on the LXX. The lexicographical description of these post-Septuagintal
sources should be separated from the lexicographical description of the
LXX.
2. When LSJ quotes €pwtijoal tiva ele elpryyny, its description is
marked as ‘Hebraism in LXX.’ Although incomplete, this information
is helpful for the reader. In many other instances, however, LSJ does not
describe satisfactorily the Septuagintal background of those words
which received a special meaning in the LXX, especially Hebraisms.
See, e.g., s.v. &6€a, €Eonodoyéouat, TpoonAuTos, xetdppoug (cf. Tov,
“Greek Words”*).
The analysis of a second Hebraism, tpoojAvtoc,!s and its biblical
equivalent 7) is complicated by the change in meaning of the latter in
the postbiblical period. In the Bible 1) denotes the ‘stranger’ and
‘sojourner,’ but in postbiblical times it was used as ‘someone who joined
the religion of the Israelites,’ especially in the phrase ?7x 7: (cf. also
the Aramaic x11, ‘proselyte’). The Greek translators, in accordance
with the linguistic reality of their own times, represented 71 almost
exclusively with tpooyAuTog, a word which apparently was coined to
denote the special meaning of 73 in postbiblical times. Consequently,
tpoarj\uToc of the LXX was bound to misrepresent many occurrences of 1).
While sometimes 71 could conceivably be rendered by tlpoo} Autos,
especially in the Priestly Code, its inappropriateness is felt particul-
arly in a verse such as Exod 22:20 a-739 prNa ona a 7D - Te yap
13 For studies on this lexeme, cf. the bibliography mentioned by Bauer, Worterbuch and
ThWNT, both s.v. See especially W.C. Allen, “On the Meaning of tpoonAvtog in the
Septuagint,” Expositor 4 (1894) 264-275 and J.A. Loader, “An Explanation of the Term
Proselutos,” NT 15 (1973) 270-277.
90 CHAPTER SIX
TpoonAvTot év yj Al yutpw. In this verse the Israelites are called
‘sojourners in Egypt,’ but in the LXX they are, as it were, ‘proselytes in
Egypt’ (similarly Lev 19:34 and Deut 10:19). The lexicographer wonders
whether there was a separate meaning ‘stranger’, ‘sojourner’ for
TPOOHAUTOG, as suggested by LSJ on the basis of evidence from the LXX.
It seems that such a meaning cannot be supported by evidence from the
LXX because within that translation tpoor\Autoc was merely a symbol
for 11. This view is based on renderings such as Exod 22:20, analyzed
above.
In the preceding paragraphs some lexicographical implications of
the use of stereotyped equivalents in the LXX have been elaborated
upon. We believe that if a certain Greek word represents a given
Hebrew word in most of its occurrences, it has become almost by
implication a mere symbol for that Hebrew word in the translation.
Thus, if a lexeme as tatrjp represents 2x almost exclusively, its
lexicographical description could be identical to that of the Hebrew
word, because it follows its Hebrew equivalent in all its meanings and
usages. Similar conclusions could be drawn with regard to many of the
standard equivalents of the LXX.!4 Thus vyq follows wn1. uldc follows
]2, even in such combinations as 1 Sam 26:16 nin 712 - uLol @avatibcewc,
56€a follows 7125, etc. Sometimes a Greek word equals a Hebrew word
only in certain translation units; see, e.g., the different equivalents of
MNxIa¥ ‘7 in the Prophets (mainly tavtokpdtwp).
The point of departure in this section was an investigation of the
lexicographical implications of lexical Hebraisms. Many such
Hebraisms resulted from stereotyped representations of Hebrew words.
Tov, “Greek Words” further elaborates on Stkalwua and related words.
The lexicographical implications of etymologizing renderings,
forming a special group of Hebraisms (cf. Tov, “Understand”* and TCU,
172-180), are analyzed next. In the sections ascribed to kaige-Th as
well as in the “LXX’ of Ruth (probably ascribed to the same revision;
14 Similar views have been expressed by several scholars with regard to individual
words: L. Prestige, “Lexicon of Patristic Greek—Hades in the Greek Fathers,” JTS 24 (1923)
476: ‘In both LXX and N.T. the precise sense of d5n¢ varies as does the particular
conception of )1Xw in any given passage.’ C. Mohrmann, “Note sur doxa,” in Festschrift A.
Debrunner (Bern 1954) 322: ‘On peut dire que tous les sens dont kabod est susceptible se
trouvent dans 8é€a et que, d’autre part, §6€a dans les LXX n’a jamais un sens étranger a
ceux de kabod.’ N.M. Watson, “Some Observations on the Use of S:katéw in the
Septuagint,” JBL 79 (1960) 266: ‘Our conclusion is that the LXX translators intended
5uxatdw to carry substantially the same range of meanings as that carried by "1x0, and
that, when they used the Greek verb, they did have the picture of a judge as clearly in
their minds as did the authors of the Hebrew Bible when they used the Hebrew
equivalent.’
THREE DIMENSIONS 91
see Barthélemy, Devanciers, 47), -1v was rendered by lkavdc. This
rendering is based on the interpretation of >7w as se-day, that is, ‘He
who is sufficient,’ an etymological conception which is known also from
rabbinic sources. Consequently, if this Greek rendition closely follows a
certain interpretation of the Hebrew, conversely that interpretation
must be taken as a source for explaining the meaning of the Greek word.
We must therefore ascribe to ikavéc that meaning of the Hebrew word
which the translator had in mind and not the one which we ascribe to
the Hebrew word. Hence, \kavéc in the LXX of Ruth does not mean ‘the
Almighty,’ as in LSJ, s.v.,!° but ‘He who is sufficient-competent.’
3. The LXX and pre-Septuagintal meanings
The lexicographer of the LXX attempts to grasp the intention of the
translators because only that intention determines the meanings of
words in the LXX. This understanding was applied to lexical
Hebraisms with the suggestion that the meaning of such words is
identical to the Hebrew word they represent. The main dimension of
LXX lexicography thus pertains to the meanings of the words in the
LXX, followed by the dimension of meanings applied to them in the
post-Septuagintal literature. However, beyond these two dimensions
the descriptions must be expanded to include the pre-Septuagintal
meanings of LXX words. This earlier dimension will be demonstrated by
returning to some of the words which were discussed above.
According to the preceding analysis, the lexical meaning of
TavtToKpdtwp in Jeremiah and the Minor Prophets is nixax (7) because it
reflects only this Hebrew phrase (114 x) and no other renderings of this
phrase are used in these books. However, this lexicographical descrip-
tion does not do justice to the Greek word because there is more to
TavTokpdTwp than just mx2ax (7). The Greek word had a meaning of its
own before it was used in the LXX and this meaning must have
influenced the translator(s) when they decided to use it as an
equivalent of mxax (‘7). The Hebrew phrase is usually translated as
‘Lord of hosts’ and these ‘hosts’ were interpreted differently in biblical
15 The English translations of the LXX translate this lkavéc in Ruth as ‘Almighty’ and
‘the Mighty One.’ See The Septuagint Version of the Old Testament, with an English Translation
by Sir Launcelot Lee Brenton (London, no date); The Septuagint Bible ... in the Translation of
Charles Thomson ... as Edited, Revised and Enlarged by C.A. Muses (Indian Hills, CO 1954). For a
correct interpretation of lkavéc, see Jerome’s commentary on Ezek 3:10 (PL, XXV, 102). See
further D.S. Blondheim, Les parlers judéo-romans et la Vetus Latina (Paris 1925) 3-15; Reider,
Prolegomena, 152, and the literature listed there.
92 CHAPTER SIX
and modern times.!© The Greek word is normally translated as
‘omnipotent’ and hence the translation equivalent reflects the
translator’s view of the Hebrew phrase. In order to do justice to the
background and use of tavtoxpdtwp in the LXX, one has to describe, i.a.,
its use in other parts of the Hellenistic world, when it was applied to
other deities.1”7 This analysis implies that tavtokpdtwp in the LXX
must be viewed at two different levels. The first level or dimension
records the background of the lexical choice Mx2x (’n) - Tavtokpdtwp by
the translators. An analysis of the meaning of tavtoxpdtwp at this
stage takes into consideration the etymological background of the
Greek word, its use outside the LXX, and possibly also the translator’s
exegetical motivations when using this word as an equivalent of (71)
mixx. The second level or dimension refers to the stage when the word
came to be used in the LXX as the stereotyped equivalent for Mx2x (‘7) in
Jeremiah and the Minor Prophets. At this stage the meaning of the
Greek has to be expressed as nix2¥ (7) since it represented that word in
all its usages.
A similar analysis should be applied to d\\ddudoc in the LXX. The
main meaning of this word at the second level is manifest since it
renders almost exclusively *nw?5 in the LXX from Judges onwards. The
basic meaning of the Greek word at the first level is also apparent: ‘of
another tribe,’ ‘foreign’ (indeed, it rendered "721 and 72: 73 in Isa 2:6 and
61:5). The lexicographical description of the first level of d\Ad@vAog is
somewhat complicated because of our uncertainty with regard to the
specific interpretation of “nw? and d\AdgvA0¢ which brought about the
present translation equivalent.18
The lexicographical description of the standard equivalence n-72 -
S.a6ykn is more complicated than the previous examples because
5.aGyKn in the LXX does not reflect the most frequent meaning of that
word, i.e. ‘testament.’ Taking into consideration the implications of the
Pentateuchal n-72 between God and his people, the first dimension of
Sta8r\kn may be reconstructed as ‘a unilateral agreement (cf. ‘testa-
16 See B.W. Anderson, IDB (N.Y. 1962) s.v. ‘Host of Heaven’ and the literature quoted
there.
17 See W.H. Roscher, Ausfithrliches Lexicon der griechischen und rémischen Mythologie
(Leipzig 1897-1909), s.v.; W. Michaelis, TRWNT II, 913-914; C.H. Dodd, The Bible and the
Greeks (London 1935) 19; B. Lifshitz, Donateurs et fondateurs dans les synagogues juives (Paris
1967) 28-29; C. Dogniez, “Le Dieu des armées dans le Dodekapropheton,” in: Taylor, IX
Congress, 19-36.
A few possibilities are analyzed by R. de Vaux, “Les Philistins dans la Septante,”
Festschrift J. Ziegler (Wurzburg 1972) 185-194. Even if de Vaux’s own suggestion is correct, it
is likely that the resemblance between dA\6 gvAot and pvAtotietpy (the equivalent of o-nw?5
in the Hexateuch) somehow influenced the lexical choice.
THREE DIMENSIONS 93
ment’) with strong bilateral overtones.’1? The second dimension of
Siabyjx7 is fully identical to n-42 whose usages it follows almost
exclusively. As a result of the stereotyped representation in the LXX,
the LXX use of &ta6¥\Km often does not suit its meaning in secular Greek.”
The distinction between the Septuagintal and pre-Septuagintal
meanings of words is relevant to LXX lexicography. It has been applied
to a few stereotyped renderings and to Hebraisms, so that two different
levels could be distinguished. These two dimensions may be recognized
in many lexical Hebraisms, and since a large part of the words in the
LXX belong to this category, the distinction pertains to many words in
the LXX.
4. The post-Septuagintal literature
Returning to the dimension of meanings of LXX words attached to them
in the post-Septuagintal period, we note that LXX lexicography must
disregard these later developments, but nevertheless the background of
this third dimension should be analyzed as a necessary step in the
understanding of LXX lexicography.
Many a word in the LXX was understood by later generations in a
way different from that intended by the translator(s). The examples
analyzed above referred to Greek words which were polysemous at the
time of the LXX, but in other cases the added layer pertains to
meanings which were created in the LXX itself.
The texts in which one searches for examples of the above-
mentioned type are the New Testament and the writings of the Church
Fathers, two sources which depended to a great extent on the LXX. The
post-Septuagintal use of €€opodoyéovat may serve as an example.
Before the time of the LXX this verb was used mainly as ‘to confess,’
and as such it was used in the LXX as a translation equivalent of A707.
However, the Hebrew verb denotes not only ‘to confess,’ but also ‘to
thank,’ and several translators who did not recognize the latter
meaning, Hebraistically rendered both meanings of 1710 by €€opodo-
yéouar. According to the previous analysis, the lexical meaning of this
€Eopodoyéopar in the LXX has to be expressed as 1717. The non-Greek use
of €EopodAoyéopat in the LXX as ‘to thank’ resulted from the artificial
19 Cf MM, s.v. “... Siafh\xT| is properly dispositio, an “arrangement” made by one party
with plenary power, which the other party may accept or reject but cannot alter.’ See also
the subsequent discussion of this word in MM. For bibliography on S:a6}\kn in the LXX, see
ThWNT and Bauer, Worterbuch, both s.v.
20 See, e.g., the treaty between Abraham and the Philistines described in Gen 21:27 as
Kal &:€6evto duddtepor S:abyxny and that of Israel’s enemies described in Ps 82(83):6 as
KaTd aov S:aby\kny S.ébevto.
94 CHAPTER SIX
nature of the translation language and was not used as such in the Greek
language. However, when the special meaning of ¢€opodoyéovat was
quoted from the LXX and used outside the framework of that
translation, it became part and parcel of the Greek language. For a
detailed analysis, see Tov, “Greek Words.”*
Likewise, the choice of odp€ as the main equivalent of 1w.2 was
natural because of their close meanings. However, the Greek word was
used also for Ww2 when denoting ‘body’ and even in the phrase 1Ww1 ¥5 -
‘all living beings.’ Consequently, while the basic meaning of odp€ in
the LXX was ‘flesh,’ its main lexical meaning should be expressed as
wa. At a third level the Hebraistic odp& - ‘body’ and doa odpe - all
living beings’ were used in the New Testament and in the writings of
the Church Fathers on the basis of the LXX.
Of the LXX words which have been introduced for the first time in
their new, ‘biblical,’ meaning in post-Septuagintal contexts, we may
mention the following words in the New Testament:2! &&eddoe (‘fellow
man’), &6€a (‘honor’, ‘glory’),22 26vn (‘other nations beside Israel’),
ETLOKETITOHAL (‘to care for’), €pwrdw el¢ elpyvny (‘ask after [a person’s]
health = ‘greet’, ‘salute’).29
The distinction between the Septuagintal and post-Septuagintal use
of biblical words is important, especially with regard to words and
usages which were unknown in the Greek language before the time of
the LXX. Such a distinction is often lacking in the entries in LSJ. For
example, é€oporoyéopat is often used as ‘to thank’ in post-Septuagintal
sources on the basis of the LXX. However, when LSJ quotes for the
meaning ‘make grateful acknowledgements, give thanks, sing praises’
evidence from the LXX, Philo, and the New Testament, the notation is
correct for the latter two, but has to be refined for the LXX.
The distinction between three different dimensions of lexicograph-
ical description will be to the benefit of LXX scholarship: the meaning
of the words in the pre-Septuagintal stage, the meaning in the LXX
itself as intended by the translators,2+ and the meaning of the words as
quoted from the LXX.
21 See especially H.A.A. Kennedy, Sources of New Testament Greek or the Influence of the
Septuagint on the Vocabulary of the New Testament (Edinburgh 1895).
a2 Cf. especially C. Mohrmann, “Note” (see n. 14 above).
23 Cf. H.St.J. Thackeray, “A Study in the Parable of the Two Kings,” JTS 14 (1912-3)
389-399 on Luke 14:31.
For the complexity of this analysis, see J.A.L. Lee, “Equivocal and Stereotyped
Renderings in the LXX,” RB 87 (1980) 104-117; see further Muraoka, “Septuagint Lexicon.”
CHAPTER SEVEN
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT
The following analysis, originally prepared in 1975, describes the need
for a LXX lexicon and analyzes some theoretical and practical problems
connected with LXX lexicography. For previous discussions, see Kraft,
Lexicography; M. Silva, “Describing Meaning in the LXX Lexicon,”
BIOSCS 11 (1978) 19-26; Muraoka, “Septuagint Lexicon”; id., Melbourne
Symposium; and the introduction to id., A Greek-English Lexicon of the
Septuagint (Twelve Prophets) (Louvain 1993).
1. The need for a lexicon of the LXX
a. Importance of the LXX for biblical research
Among the textual witnesses of the Bible, the LXX is the most
important source for the recognition of Hebrew readings that differ
from MT. The majority of these variant readings are of importance for
the textual criticism of the Bible, but some bear also on literary
criticism (see, Tov, “Jeremiah”*).
The LXX also reflects much important information about the exegesis
of its translators. Some books are of particular importance in this
regard since their exegetical traditions reveal much about the cultural
and intellectual background of their translators (especially Isaiah, Job
and Proverbs; see Tov, “Theology”*). The understanding of these
exegetical traditions depends much on the correct analysis of
translation techniques and the translators’ lexical choices.
b. Importance of the LXX for the intertestamental literature
The LXX is not only a translation of a corpus of Hebrew literature, but
also a source for later literature, for the translators of the LXX created
a vocabulary of translation Greek which had a great influence on
subsequent Jewish Greek literature, both translations and compositions
originally written in Greek: Philo, Josephus, Jewish Greek historical,
exegetical, poetical and apologetical writings (collected by Reinach in
96 CHAPTER SEVEN
1895, Denis in 1970, and Stern in 1976)! and the so-called Pseudepi-
grapha (e.g., the Testaments of the Twelve Patriarchs, the Testament
of Abraham, the Testaments of Job, Joseph and Asenath, etc.). Many
words in these compositions cannot be described properly without
reference to the vocabulary of the LXX, as has been recognized by many
scholars, e.g., R.H. Charles, The Greek Versions of the Testaments of
the Twelve Patriarchs (Oxford 1908) xl ff.; M. Delcor, Le Testament
d’ Abraham (Leiden 1973) 28 ff.; Daniel, Recherches, 375-379; H.G.
Meecham, Aristeas (Manchester 1935) 52-77, 316 ff.
c. Importance of the LXX for New Testament research
The language and vocabulary of the LXX are an important source for
understanding the language and meaning of the New Testament.
Several of the writings of the New Testament, in particular the
Gospels, and among them especially Luke, were written in a special
type of Greek which was once characterized as biblical Greek, Jewish
Greek or the “tongue of the Holy Spirit,’ but whose special character is
now recognized as largely due to its dependence on the language of the
LXX. The degree of this dependence is the subject of much debate;
however, it is probably agreed by most scholars that both the language
of the LXX and a Semitic influence or source (written or oral) of some
kind were instrumental in the creation of the peculiar language of the
New Testament. Some scholars have attempted to distinguish in the
Gospels between Septuagintalisms and the influence of this Semitic
source (see especially H.F.D. Sparks, “The Semitisms of St. Luke’s
Gospel,” JTS 44 [1943] 129-138) and in this way the influence of the LXX
on the New Testament can be described more efficiently. The authors of
the Gospels often imitated the language of the LXX. Consequently they
often used ‘Greek words and Hebrew meanings’ (the title of D. Hill’s
book of 1967, dealing with the Septuagintal background of the
vocabulary of the New Testament [see Tov, “Greek Words”*]) and
consequently the lexicography of the New Testament depends much on
that of the LXX. Lexicographers of the New Testament have always
paid much attention to the vocabulary of the LXX and lexica such as
those of Thayer (New York/Edinburgh 1886), Preuschen (Giessen 1910),
Abbott-Smith (3rd ed.; London 1937) and Bauer, Wérterbuch (1963)
often mention the Hebrew words which are reflected in the New
17. Reinach, Textes d’auteurs grecs et romains relatifs au Judaisme (Paris 1895); A.-M. Denis,
Introduction aux Pseudépigraphes grecs d’Ancien Testament (SVTP 1; Leiden 1970); M. Stern,
Greek and Latin Authors on Jews and Judaism, I-III (Jerusalem 1976-1984).
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT 97.
Testament through the intermediary stage of the LXX. The
Septuagintal background of these words is a significant element for the
lexicography of the New Testament and in this area a lexicon of the
LXX would play an important role.
d. The lack of a lexicon of the LXX
The LXX is of importance for the study of not only the Bible, the
intertestamental literature, and the New Testament, but also of the
sources which depend on the LXX such as the Church Fathers and the
translations made from the LXX.
This absence of a LXX lexicon is felt especially in view of the
relatively large number of adequate tools which are available for the
study of the LXX and in view of the existence of lexica in related areas.
The tools which are available to the student of the LXX are relatively
numerous in comparison with those which are available for the study
of the other versions of the Bible. LXX scholars resort to a relatively
good Greek-Hebrew concordance (HR, 1897; revised edition, 1998),
including a reverse index by T. Muraoka, to often thorough though
incomplete grammars (Helbing, Grammatik, 1907; Thackeray, Gram-
mar, 1909), and to excellent critical editions (G6éttingen Septuagint,
1931- ; Cambridge Septuagint, 1906-1940, both incomplete; Rahlfs,
1935). The number of critical studies on various aspects of the LXX is
extremely large. The Classified Bibliography (1973) contains 201 pages
of bibliographical references for the period between 1860-1970, while
Dogniez, Bibliography covers the period between 1970 and 1993 in 329
pages. In view of the broad coverage of the LXX, a lexicon remains a
desideratum. Since this article was published in 1975, a full-size
lexicon on the LXX of the Minor Prophets was published by T. Muraoka,
A Greek-English Lexicon of the Septuagint (Twelve Prophets) (Louvain
1993)—see the review by J. Lust, “Two New Lexica of the Septuagint
and Related Remarks,” JNSL 19 (1993) 95-105. Furthermore, two concise
lexica were published, the second one of which is especially helpful: F.
Rehkopf, Septuaginta-Vokabular (Gottingen 1989); J. Lust, E. Eynikel,
and K. Hauspie, A Greek-English Lexicon of the Septuagint, I-II
(Stuttgart 1992, 1996).
e. Available lexical tools
Three older dictionaries of the LXX are available of which two are
hardly known and of little use: those of Rosenarch (1624) and Ewing
(3rd ed.; 1827) list merely one or two equivalents for the words of the
LXX, without any further description or biblical references. The third
98 CHAPTER SEVEN
one, on the other hand, is well known and frequently referred to as ‘the’
lexicon of the LXX: Schleusner, Thesaurus. However, it, too, is of
limited use. Schleusner’s lexicon does not resemble other lexica, because
it does not concentrate on the description of the words in the LXX, but on
their Hebrew equivalents which are translated into Latin. The lexicon
is probably more helpful as a concordance (unlike HR, it provides the
Hebrew equivalents of the ‘Three’) and as a storehouse of remarks on
the translation technique and text of the LXX than as a lexical tool. It
was published long before the ground-breaking studies of Deissmann
(1895-1910) on the close relationship between biblical Greek and
Hellenistic papyri from Egypt. Consequently neither Deissmann’s
approach nor the new data themselves are reflected in Schleusner’s
Thesaurus. The lexicon was written in Latin so that it is not used much
by modern scholars.
LSJ, together with its Supplement by E.A. Barber (Oxford 1968) and
the Revised Supplement by P.G.W. Glare (Oxford 1996), should not be
omitted from a review of lexical tools for the study of the LXX. This
lexicon is used by most biblical scholars because it constitutes the best
available lexical source for the Greek language. This dependence is
justifiable as long as LSJ is taken as a general source of information for
the Greek language, and not for its remarks on the meanings of words in
the LXX—unfortunately LSJ contains many errors with regard to the
LXX. Its most frequent methodological error is that LSJ ascribes to many
LXX words the meaning of their Hebrew Vorlage, even when the
translator’s consonantal Vorlage presumably differed from MT. A good
example is auvSecpdés (‘conspiracy’) for which LSJ (s.v. V) created a
new meaning ‘sodomy’ because it represents v7? (‘male prostitute’) in 1
Kgs 14:24. However, in this verse the translator read wv? (ie.,
‘conspiracy’) instead of MT (by way of metathesis and interchange of
dalet/resh) and hence no new meaning should be posited for the LXX.
For examples of similar mistakes in LSJ, see G.B. Caird, “Towards a
Lexicon of the Septuagint, I, IL,” J/ThSt NS 19 (1968) 453-475; 20 (1969)
21-40; J.A.L. Lee, “A Note on Septuagint Material in the Supplement to
Liddell and Scott,” Glotta 47 (1969) 234-242; R. Renehan, “Greek
Lexicographical Notes: Fifth Series,” Glotta 50 (1972) 38-60; id., Greek
Lexicographical Notes. A Critical Supplement to the Greek-English
Lexicon of Liddell-Scott-Jones (Untersuchungen zur Antike und zu ihrem
Nachleben, Hypomnemata 45, 74; Gottingen 1975, 1982).
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT 99
f. Modern lexicological studies
LXX lexicography is aided much by the many lexicological studies
written in this century on words and word groups in the LXX, the New
Testament, or both. The approaches of such studies vary greatly. One
encounters mere concordance studies describing the statistical aspects of
a Greek-Hebrew equivalence, studies dealing with the background of a
lexical equation, studies which treat the history of a word in the Greek
language, studies which are interested mainly in Hebraisms, and
statistical studies comparing the vocabulary of the various books of the
LXX, etc. Few studies deal with all the aspects needed as background
information for compiling an entry in a lexicon of the LXX. Of the latter
type, the following studies should be mentioned in particular: Da
Fonseca on &tayKn (1927-28), Repo on pra (1951), Paeslack on otdéw,
etc. (1953-54), Daniel on cult terminology (1966) and Monsengwo
Pasinya on vépyos(1973) and the studies included in Muraoka, Melbourne
Symposium. For bibliographical references, see E. Tov, Lexical and
Grammatical Studies on the Language of the Septuagint (2d ed.;
Academon: Jerusalem 1975); and the aforementioned LXX biblio-
graphies.
2. For whom 1s the dictionary intended?
The lexicon is to be used by scholars specializing in the Hebrew Bible,
intertestamental literature, New Testament, Patristics, Jewish Helle-
nism, and Greek linguistics. The ideal reader of the lexicon would have
a good knowledge of both Hebrew and Greek. As indicated in sections
Sba, many words in the LXX follow their Hebrew equivalents in all
their meanings; consequently within the LXX the lexical meaning of,
e.g., d:aOr\kn has to be expressed as n°12, of which it is always a
mechanical equivalent. However, the reader of the lexicon of the LXX
cannot be expected to examine the meanings of n°-72 in a lexicon of the
Hebrew Bible and therefore the lexicon of the LXX must provide them.
A detailed treatment of 5ta@{xn is needed also because the nature of the
lexical choice must be illustrated by the various types of S:a8}Kn in the
LXxX.
The planned dictionary should provide all the necessary lexical
information, as in Bauer, Wérterbuch relating to the New Testament,
and will not presuppose other lexica (as Lampe’s dictionary of Patristic
Greek does with regard to LSJ).
100 CHAPTER SEVEN
3. Some theoretical issues
a. The language of the LXX in the translated books
The planned dictionary records the language of the LXX; it is a
linguistic tool and it is therefore in order to dwell somewhat on the
nature of that language.
The special character of the language of the LXX may be described
in various ways, as has been done in the past, as an exponent of the
Greek of Hellenistic Egypt, as a Jewish Greek dialect, or as translation
Greek. The various positions on the nature of ‘biblical Greek’ have been
described by J. Ros, De studie van het bijbelgrieksch van Hugo Grotius
tot Adolf Deissmann (1940) and J. Vergote, “Grec, biblique,” DBSup 3
(1938) 1320-1369.
We adhere to the view which ascribes the special nature of the
language of the LXX in the first place to its background as a translation.
Doubtlessly, certain of its special features are due to the fact that the
LXX reflects the Egyptian branch of Hellenistic Greek, but this
situation accounts only for some idiosyncrasies of the language of the
LXX.
Finally, with regard to the possibility that the Greek of the LXX is
a typical exponent of a Jewish Greek dialect, the existence of a Jewish
Greek dialect cannot be substantiated in any one period. The assumption
of such a dialect must be distinguished from a Jewish Greek vocabulary
containing mainly technical ‘Jewish’ terms which may be posited in the
time previous to the translation of the Torah. It must also be
distinguished from the influence emanating from the vocabulary of the
LXX on subsequent literature (see 1c).
b. The aim of LXX lexicography with regard to the translated books
A lexicographer analyzes words in languages and literatures with the
ultimate aim of describing their meanings in a dictionary. This task is
not easy with regard to ancient languages and literatures where no
informants can be consulted. Equally difficult is the lexicographical
description of a translation because the language of a translation is
often unnatural. These two difficulties are combined in the
lexicographical description of an ancient translation—in our case, the
LXX. In abstract terms, the lexicography of a translation aims at
recovering the meanings of the words in the translation as intended by
the translator(s), and not by the post-Septuagintal literature (see Tov,
“Dimensions”*).
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT 101
c. The nature of the canon of the LXX
The canon of the LXX contains books of a varied nature, early and late,
an official translation (Torah) and enterprises of individuals, original
translations as opposed to revisions. Such revisions are contained, for
example, in the “LXX’ of Ecclesiastes (Aquila), in parts of Samuel-
Kings (kaige-Th), in Daniel (Theodotion?), in Ruth (kaige-Th).
The recognition that the LXX is an amalgam of different translation
units has repercussions on the work of the lexicographer. Since the LXX
consists of many different units one encounters a larger number of
separate meanings than would have been the case if the canon of the
LXX had been more homogeneous. This situation may be illustrated by
the use of \kavdc—see Tov, “Dimensions”*, p. 91. See further below, 5a.
d. Translation equivalents
Because a lexicon of the LXX describes a language which is mainly
translation Greek, the lexemes to be described are not simply words,
but, as a rule, they are translation equivalents, too. It is this dimension
of the language of the LXX which makes LXX lexicography a
challenge.
The lexicographer has to bear in mind that many words in the LXX
were meant to represent their Hebrew equivalents faithfully. There-
fore one must constantly pay attention to the linguistic aspects of the
lexical equations of the Hebrew (Aramaic) and the Greek. E.g., one
must realize that certain equivalents are based on the Aramaic rather
than the Hebrew root, that a certain rendering imitates the sound of
the Hebrew, that another rendering reflects a certain shade of the
Hebrew which would not have been clear solely on the basis of the
Greek, and that in yet other cases a Hebrew word is always represented
in the LXX by the same equivalent. All these aspects of translation
technique must be taken into consideration as part of the lexico-
graphical description, while the purely descriptive characterizations
as ‘free rendering,’ ‘theological,’ or ‘faithful’ form no part of the
lexicographical description (against Schleusner, Thesaurus).
4. The scope of the lexicon
The desirable scope of the LXX lexicon has been discussed in the studies
included in Kraft, Lexicography, esp. in Kraft’s own contribution, pp. 31
ff. This discussion is continued here, taking as point of departure
Schleusner, Thesaurus. Schleusner produced a biblical lexicon which
102 CHAPTER SEVEN
covers the canonical books of the LXX (both translations from Hebrew
and Aramaic and compositions originally written in Greek), the
‘Three,’ and the Apocrypha. Such a lexicon could be expanded to
include the Pseudepigrapha and other Jewish Greek sources, for these
compositions often resemble the Apocrypha which are covered by
Schleusner. A future dictionary could also be enlarged with a view
toward encompassing all Jewish Greek sources. On the other hand, a
smaller scope than that of Schleusner’s lexicon can also be envisaged.
The possible shapes for a LXX lexicon are compatible with different
aspects of the LXX literature:
1. A larger lexicon of Jewish Greek sources. A lexicon of this scope is
based on the understanding that the LXX forms part of the Jewish
Greek literature which should be described en bloc. Such a lexicon
could cover the period from the third century BCE until the third or
fourth century CE. The sources to be covered would be the canon of
the LXX, the ‘Three,’ the Pseudepigrapha, literary compositions
(historical, exegetical, poetical and apologetical) collected by
Reinach (1895), Denis (1970), Stern (1976), papyri and inscriptions.?
The writings of Philo and Josephus should probably be covered by
this lexicon, but they might be excluded on practical grounds.
2. A smaller lexicon of Jewish Greek sources, similar to (1), but
covering only the canon of the LXX, the ‘Three,’ and the
Pseudepigrapha.
3. A complete lexicon of biblical Greek, similar to (2), but omitting the
Pseudepigrapha and covering all extant translations of the Bible in
Greek.
4. A smaller lexicon of biblical Greek, covering the canon of the LXX
and not the ‘Three’.
5. A lexicon of Jewish Greek translation literature surrounding the
LXX. Although the Alexandrian canon contains compositions which
were originally written in Greek, it forms at the same time the most
important collection of works in translation Greek. A lexicon which
would cover the translation character of the LXX could also cover
some of the Pseudepigrapha which were translated, as was the
LXX, from Hebrew and Aramaic, and whose vocabulary is often
similar to that of the LXX. A lexicon of this type may or may not
cover those sections of the Alexandrian canon which were not
translated from Hebrew and Aramaic.
2 For references, see n. 1. For Jewish-Greek inscriptions, see M. Tcherikover, A. Fuks, and
M. Stern, Corpus Papyrorum Judaicarum, {-II] (Cambridge, MA 1957-1964).
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT 103
All five types of lexica outlined above serve scholarly needs. The scope
of the biblical lexica (possibilities 3 and 4) is better demarcated than
that of the other forms since it covers the Alexandrian canon with or
without additions. This scope may therefore be more desirable than
that of the other possibilities even though the biblical material itself
is heterogeneous. We now tum to other aspects of the lexica which are
not merely biblical.
The wide scope of the larger lexicon of Jewish Greek sources (1) is
justifiable not only because the compositions to be covered are of less
direct relevance for LXX research, but also because many of the Jewish
Greek sources have elements in common with the LXX. However, some
Jewish Greek sources are of less direct relevance for LXX research
because they lack dependence upon the LXX (some contain Jewish
technical terms that are not reflected in the LXX). Thus for the student
of the LXX, the Hebrew Bible and the New Testament a larger lexicon
of Jewish Greek sources is not a necessity. The wide scope of this lexicon
may complicate the annotation, but on the other hand such a lexicon
will contain details which may provide background information for the
lexical choices of the LXX.
The smaller lexicon of Jewish Greek sources (2) is a biblical lexicon
with a difference, for it also covers the Pseudepigrapha. The shape of
this lexicon would not be easy to determine because what is included in
the term “Pseudepigrapha’ is subject to debate.
A lexicon of translation Greek (5) is attractive, but its scope creates
more problems than it solves. If such a lexicon excludes part of the
Alexandrian canon in order to limit the scope of the work to that of
translation literature, many would find this lexicon too narrow. On the
other hand, if part of the canon is not excluded, the lexicon would be too
heterogeneous. Since the extra-Septuagintal translation literature
belongs to the Pseudepigrapha, a lexicon of translation Greek may, in
fact, also be realized in the form of possibility (2).
Much may be said in favor of a dictionary of translation Greek
because of the common vocabulary of many of the sources covered.
However, the lexicographical description of the Pseudepigrapha (and
much of the Apocrypha) can never be complete, for the Semitic
Vorlagen of these books have been lost in most cases, and the inclusion
of the Pseudepigrapha within the group of translated books therefore
causes more practical problems than the ones created by the
Apocrypha, which are part of the canon of the LXX. A more serious
argument against the composition of a lexicon of translation Greek is
104 CHAPTER SEVEN
the fact that the original language of many books of the Apocrypha
and Pseudepigrapha has not been determined.
In short, several arguments may be raised against possibilities (1),
(2) and (5). In addition, these lexica do not cover a relatively well
defined corpus such as the Alexandrian canon. A purely biblical lexicon,
therefore, may be more advisable, although such a lexicon should
necessarily include references to extra-biblical Jewish Greek literature.
The preference for a biblical lexicon is based partly on the fact that
its borders are determined by those of the Alexandrian canon (this view
has been accepted in the concise lexica by Rehkopf and Lust/Eynikel/
Hauspie mentioned above). The precision of this demarcation should
not be overly stressed because the Alexandrian canon has different
shapes in different sources, both in the various manuscripts and in
descriptive lists of its contents. However, the uncertainty with regard
to these borders should not deter us from using some form of the canon as
a criterion for the sources covered by a lexicon. One could opt for an
expanded form of the canon (such as the one reflected by codex A) ora
more limited one (such as the canon reflected by codices B and S). The
editors of the Gottingen LXX had to make similar decisions when they
decided to include certain books in their collection of text-editions,
while excluding others. Furthermore, a decision of this kind is less
arbitrary than trying to determine which of the Pseudepigrapha
should be included, or which books may be considered translations and
hence ought to be covered in the dictionary of translation Greek (5).
The actual shape of a biblical lexicon varies with the different
opinions concerning the later Greek translations, viz. the ‘Three’ and
the revisions of Origen (the Hexapla) and Lucian. The situation of the
latter two differs from that of the former three. Let us first discuss the
‘Three’.
If the shape of a biblical lexicon were determined by the contents of
the Alexandrian canon, as suggested above, the ‘Three’ should be
excluded. Although the revisions of the ‘Three’ are based on
translations included in the Alexandrian canon, they were not meant
themselves to be included in that canon. On the contrary, the revisions
of Aquila and kaige-Th as well as others were prepared as a reaction
against that canon. The translations of the ‘Three’ should therefore not
be covered systematically in a lexicon of the LXX (4), but rather in an
expanded lexicon of biblical Greek (3). Two practical arguments further
support possibility (4):
a. It is difficult to provide an adequate lexicographical description
of the words used by Aquila and kaige-Th. These two revisers did not
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT 105
produce a translation in any traditional sense of the word, but rather
used Greek word-symbols to represent the corresponding Hebrew and
Aramaic words. These symbols were, as a rule, ordinary Greek words,
but often they were unintelligible to the uninitiated reader who had no
access to the Hebrew text. Thus in Gen 1:1, €v xkedadalw Exticev Beds
ovv TOV OvVpavov Kal ouUY THY yAV, the meaning of Aquila’s ovv is
expressed best by its Hebrew counterpart, nx. Similarly, kedddaLov
cannot be explained satisfactorily according to any of the known
meanings of that word in the Greek language; it was chosen by Aquila
merely because it is a derivative of kedadr, aS N°WN 7 is a derivative of
wx. Hence Aquila’s lexical choice must be viewed against the
background of the equation Wx7: Mmwxn = kedadn: Kepddatov. Because of
the many lexical equations of this kind in Aquila’s revision, the
lexicographical description of Aquila’s words would probably produce
mainly an annotated list of their Hebrew equivalents which may be
translated into English. The same holds true for much of kaige-Th’s
revision. In our view, the mere listing of Hebrew equivalents does not
produce a lexicon; it produces indexes of the type of J. Reider and N.
Turner, Index to Aquila (VTSup XII; Leiden 1966).
b. Practical problems are encountered in the description of words
from the ‘Three.’ The text of the ‘Three’ has been preserved in a few
running texts, but for the most part as individual words or phrases
recorded in the margins of manuscripts or quoted by Church Fathers as
variants to renderings in the LXX. Moreover, much of this material has
been transmitted in Syriac, Latin, and Armenian translation and often
cannot be reconstructed satisfactorily into Greek. A further problem
caused by the fragmentary transmission of the “Three’ is that often the
words are transmitted out of context—hardly an ideal condition for a
lexicographer.
As a result of these practical problems with regard to the ‘Three,’ in
the initial stage a smaller biblical lexicon is preferable. Such a lexicon
would cover only the canonical books of the LXX (including the
Apocrypha). The lexicon would be based not only on the eclectic or
diplomatic text of the critical editions, but also on readings in the
critical apparatuses. It would have to be determined whether these
variants should be covered systematically, or whether certain groups of
variants ought to be excluded. This problem applies especially to the
Origenic and Lucianic readings. On the one hand, recensional readings
should not be covered systematically, for they are later in date than
the revisions of the ‘Three’ which are excluded from the smaller
biblical lexicon. The Origenic (Hexaplaric) and Lucianic manuscripts
106 CHAPTER SEVEN
belong to the transmitted text of the LXX, and should therefore be
treated as part of the ‘LXX’ in the lexicon, similar to their treatment in
the Gottingen Septuagint. Moreover, any attempt to separate the
Origenic and Lucianic elements from the LXX (the OG) would be
unsuccessful because in the course of the transmission, the revised
elements of the former have been mixed greatly with the unrevised
elements of the latter. In practical terms, all variants listed in the first
apparatus of the Gottingen Septuagint should be covered by the lexicon.
Probably an exception may be made for words which have been added
sub asterisco from the ‘Three’ in the revisions of Origen and Lucian
because the non-Septuagintal nature of these elements is evident.
In conclusion, we suggest that a future LXX lexicon should in its first
stage cover the canonical books of the LXX (that is, including the
Apocrypha) on the basis of both the text and variants contained in the
first apparatus of the Géttingen and Cambridge Septuagints, with the
possible exclusion of asterisked words. Other sources will be referred to,
but need not be covered systematically.
But also this suggestion is problematical: It excludes the ‘Three’
from a systematic coverage, while the canon of the ‘LXX’ itself contains
sections which are ascribed to the ‘Three’ (the ‘LXX’ of Ecclesiastes is a
product of Aquila’s and several sections in Samuel-Kings are ascribed to
kaige-Th). Hence, if this analysis is followed, words from the ‘Three’
will nevertheless be recorded in the smaller lexicon of biblical Greek
through indirect means. Consequently, this proposal may not provide
the basis for the ideal LXX lexicon, but it has the virtue of avoiding
other greater difficulties. Probably no form of LXX lexicon is ideal
because the heterogeneous character of the canon of the LXX does not
provide favorable conditions for any consistent solution.
5. Sources for LXX lexicography
The lexicographical description of originally Greek compositions
within the LXX does not differ from that of Greek compositions outside
that literature. We concentrate here on the lexicographical description
of the translated books which differs from that of the other books
because of the special nature of the translation.
In the search for the meanings of LXX words, the lexicographer resorts
to both internal and external evidence. Internal evidence involves
information deriving from the LXX itself or from its Hebrew Vorlage.
External evidence involves Greek extra-biblical sources. _
SOME THOUGHTS ON A LEXICON OF THE SEPTUAGINT LO7
a. Internal evidence
For a translation such as Aquila’s, internal evidence provides the best
background information for its Greek words (see above). If we take the
LXX as a whole, internal evidence is probably as important as external,
but the relation between the two sources differs from book to book
depending on the nature of the translation. In literal translation units,
internal evidence is important, but it hardly plays a role in very
paraphrastic translation units. For a description of various aspects of
internal evidence (polysemy, stereotyped renderings, etymologizing
renderings, and neologisms), see Tov, “Dimensions””; “Greek Words”*;
and TCU.
b. External evidence
(a) The recording of external evidence
The lexicon should record all external (that is, extra-biblical, secular)
evidence which is needed to establish the meaning of a word in the
LXX. Such evidence is also needed to illustrate the linguistic
background of certain Hebrew-Greek equivalents.
It is difficult to decide how much parallel material should be
recorded for the individual meanings in large entries, in particular of
words which follow all the meanings of their equivalent Hebrew word.
For example, although N.P. Bratsiotis provided extra-biblical
parallels for all meanings and usages of Wuxr, in the LXX on the
equivalence of w51 and guy in the LXX (“woi1-pux7 Ein Beitrag zur
Erforschung der Sprache und der Theologie der Septuaginta,”“VTSup 15
[1966] 58-89), he reached the conclusion that w51 and uy cover each
other rather well. Most of these parallels may be superfluous.
(8) External sources to be covered
External sources to be covered comprise in principle all Greek texts, both
literary and non-literary, early and late. Hellenistic sources are of
particular importance, especially those from Egypt. In order to cover
the Greek literature fully, all relevant dictionaries need to be
consulted, both ancient, medieval and modern (extant dictionaries are
listed by H. Riesenfeld, Repertorium lexicographicum graecum [Uppsa-
la 1953]). Of special importance for the LXX are the lexica by
Hesychius, partly based on an early biblical glossary (ed. Latte,
Hauniae 1953); Schleusner, Thesaurus; Bauer, Worterbuch; Lampe
(Church Fathers, 1968); LSJ (1940, 1968, 1996); Sophocles (late Greek,
108 CHAPTER SEVEN
New York 1900); F. Preisigke, E. Kiessling, H.-A. Rupprecht, and A.
Jordens, Worterbuch der griechischen Papyrusurkunden etc. (Berlin/
Amsterdam / Wiesbaden 1925-1991); and MM.
Close attention must be paid to Greek papyri from Egypt as these
often provide the best parallels to the vocabulary of the LXX,
especially with regard to its technical terms. E.g., the tapdSetcos used
in Genesis for 71, has been described on the basis of papyri as ‘an area of
cultivated ground containing chiefly fruit-trees, at times also other
types of tree, vines, and possibly other plants, and perhaps protected
by a wall’ (Lee, Lexical Study, 55).
6. Some remarks on the contents of the entries
In principle the treatment of words in originally Greek books should be
identical to that of words in secular Greek sources. A few words on the
translated books are in order.
The main purpose of the lexicon is to record the meanings of the
words which presumably were intended by the translator(s).
Hebrew equivalents are mentioned when the editor believes that
the Hebrew word best expresses the content of the Greek lexeme.
Accordingly, the mere use of Hebrew characters indicates a certain idea
which the lexicon wants to convey to the readers.
Some details will be indicated in a special way such as lexical
Hebraisms, LXX neologisms, and statistical details.
Sample entries were presented by Tov, “Greek Words”* and
Muraoka, “Septuagint Lexicon.” Beyond the experimental stage,
Muraoka’s lexicon of the Minor Prophets (see p. 97) contains the first
modern enterprise of LXX lexicography.
CHAPTER EIGHT
‘GREEK WORDS AND HEBREW MEANINGS’
Greek Words and Hebrew Meanings—Studies in the Semantics of
Soteriological Terms (SNTS Monograph Series 5; Cambridge 1967) by
D. Hill demonstrates that certain words in the New Testament are used
in senses which reflect their Hebraic background (through the LXX)
rather than their natural Greek heritage. In the wake of that study,
the following analysis centers on a few words which carry some senses
which have been determined by the Hebrew together with a
lexicographical description of these words in full. At the end of each
analysis a sample entry for a LXX lexicon is suggested.
1. &tkalwpa
LSJ describes the basic meaning of 6tkalwya as a legal term (I), a term
connected with the world of lawsuits. It also provides (II) the meaning
‘ordinance, decree’ for the LXX and subsequent literature.
act of right, opp. d&l{knua, Arist.; duty Philo 2.199; prop. amendment of a wrong, opp.
SikaLoTmpadynua, Aristotle; hence:
I a. judgement, penalty, Plato
b. justification, plea of right, Thucydides 1 41,1, Isocrates, Aristotle, LXX 2 Ki
19.28(29), PLond ii AD
cap pl. pleadings, documents in a suit, papyri tli B.C., also credentials pap ii A.D.
d. act of &txalwors i 3, N.T.
II ordinance, decree, LXX Gen 26.5, Exod 15.26 (pl.), al., NT
Discussion:
[I] LSJ records one occurrence of the meaning ‘justification’ for the
LXX. In the ThDNT this sense is defined. as ‘legal ground or claim,’
exemplified by 2 Sam 19:29 (listed as ‘LXX 2 Ki’ in LSJ):
2 Sam 19:29 Kal tl éotty pot ét Sikalwua Kal Tot Kekpayévat
Le ETL IPOS TOV BacLr\€a
What further right have I to cry to the king?
= P70 OX TW pyt?) APTS WY 7%? wr
110 CHAPTER FIGHT
This sense of 5ikalwya comes close to the meaning recorded for classical
Greek by LSJ (justification), but it is better defined in ThDNT with the
same example from Thuc. I 41,1 6tkatupata tdSe pds buds éxopev,
referring to the established legal claims of allies. At the same time,
the meaning of dtKalwya in 2 Samuel is rightly determined on the basis
of the Greek context as ‘justification.’
The word is also used in a similar sense as ‘cause’ or ‘case’:
2 Chr 6:35 Kal TOLYOELS TO Stkalwua avTuy
And You will handle their case.
Jer 18:19 elLadKkovoov THs duvijs Tot Skatapatds pou
Listen to the voice of my case. Cf. also 11:20.
This is the only overlapping sense between the LXX and the general
meanings of that word recorded in LSJ. The situation differs for the
ThDNT, Bauer, Worterbuch, and subsequently for BAGD, which gives
the following description:
1. regulation, requirement, commandment (so mostly LXX; Philo; Josephus; Cass.
Dio (ii-iii A.D.; POxy 1119,15 Tay é€atpétuv ts teTtépas tatplbos SikaLwpdTwy )
2. righteous deed (Arist. 1135a,12f; 1359a, 25; 1373b,1; 3 Km 3:28; Bar 2:19; N.T.).
Ss [discussion of Ro 5:16 in the New Testament]
The meaning which apparently reflects the most frequent sense of
Sixalwua in the LXX (defined as ‘ordinance’, ‘decree’ in LSJ and as
‘regulation, requirement, commandment’ in BAGD) is not documented
outside the LXX according to LSJ, but is documented according to BAGD,
[2,3] Another difference between the two lexica is that BAGD (as
well as ThDNT) adds a meaning ‘righteous deed’ for which it provides
support from Aristotle, not recorded by LSJ. Three issues are at stake:
1. Does a meaning ‘righteous deed’ exist for the LXX, as indicated
by BAGD?
2. Do the two possibly separate meanings, ‘righteous deed’ and
‘justice’ appear in the LXX?
3. Is one of these senses, or possibly both, supported by external
evidence?
We first turn to the best examples from the LXX relating to the
meanings ‘righteous deed’ and ‘justice’:
1 Kgs 3:28 dpdvyats Geod Ev avtu Tov trocetyv Stkalwya
The wisdom of God was in him (scil. Solomon) to do
justice/a righteous act.
Prov 8:20 év O601¢ SiKaLooUVnS TeptTaTi, kal dva péoov TplBwv
SLKALULLATOS GVaoTpPE PoudL
“GREEK WORDS AND HEBREW MEANINGS’ 111
I walk in the ways of righteousness, and on the paths of
justice I go about.
Prov 19:28 d éyyuuipevos Tratda ddpova kabuBplCer Stkalwya, oT6La
St doeBuiv katatletat Kplaets
He who answers for (?; becomes surety for) a foolish
child insults justice and the mouth of the impious
devours judgments.
In the latter two verses 6tkalwua has the sense of ‘justice’ in general, as
is clear from the parallel words 8tkatootvy and kplots. On the other
hand, in 1 Kgs 3:28 d.kalwpa means either ‘justice’ or ‘righteous act,’ as
one of the manifestations of justice. Such a meaning is also evidenced for
Bar 2:19, in the section which has been translated from Hebrew.
Bar 2:19 Sti ovk él Ta SiKaLwWuaTa TOV TaTEpwv Hud ...
KaTaBdAdopev Tov EdEov HWY KATA TPdOWTOV COU
For not by virtue of the righteous deeds of our ancestors
... dO we present our supplication before you.
The meanings ‘righteous act’ and ‘justice’ ina general sense are thus
evidenced for the LXX. Neither sense is listed in LSJ, while BAGD lists
the first one, supported by evidence outside the LXX.
Returning to 1 Kgs 3:28, we note that the context favors the general
meaning ‘justice’ rather than ‘righteous deed.’
[4] The frequency of the use of &ikalwya in the LXX as ‘ordinance,’
‘decree’ (LSJ) or ‘regulation, requirement, commandment’ (BAGD) is
related to its Hebrew source, supported by external evidence. This
Stkalwya is often used together with similar words, e.g.:
Gen 26:5 Kal €dvAatev Ta TpCOTAyLaTa [Lou Kal Tag EvTOAAS LLoU
Kal Ta StkatwpaTd ov Kal Ta vépLLd LOU
And he kept my commandments, my orders, my
Sikatwpatd, and my laws.
The context does not enable us to determine the exact meaning of
Sikalwpa, but the general sense is clear. In his conversation with Isaac,
God mentions the various commandments which Abraham has ful-
filled. Grouped with the évtodal, véptpia, and tpootdypata, the word
Sikalwua probably indicates something issued by God. Also in Exod
15:25, Skalwya, once again in plural, is given by God, in this case to
Moses:
éxet €8eTo avTu Sikarwpata Kal Kploets
There He gave to him &:katwyata and judgments.
112 CHAPTER EIGHT
Sikalwa is used not only in parallelism with the mentioned words, but
also in other close grammatical relationships:
Num 30:17 Tatta Ta dikatuata doa évetelAaTo KUpLOS
These are the &katwyata which God ordered.
Num 31:21 TOUTO TO SikalwpLa Tot VvdjLou
This is the Stkalwpa of the law.
The root 8tk- could imply that every dikafwyua is just or righteous, but
several of the usages of d:kalwya outside the LXX are connected with
the legal world, without being intrinsically right. Therefore a neutral
term such as ‘decree’ or ‘ordinance’ covers the Greek word well.
The fact that little evidence has turned up so far for this meaning
outside the LXX may be coincidental. The translators used this word
from Genesis onwards mainly for three Hebrew words, ?n (46 times), 7?n
(26 times), and bdwn; (41 times). It also occurs 6 times for 7375, 3 times for
ni7¥, and once each for 777, 718M, 2°97, 798, and 3°77. Usually words of the
5.k- group reflect words of the root /273, so that the equivalents of
Sikalwpa, not related to this root, are noteworthy. The choice of
Sikalwpa probably reflects a meaning common to both v5wn and apn /Pn,
viz., ‘decree.’ This sense of 8tkalwpua is evidenced also outside the LXX
(not mentioned in LSJ). Of the sources mentioned by BAGD, Philo and
Josephus are probably irrelevant, since they depend on the LXX. But
two other sources are relevant. They are later than the LXX, but are not
found in the realm of its influence, viz., Dio Cassius and P. Oxy 1119,15
(254 CE) Ta&v €Earpétwv THs NpETEpas tatpl&os Sikarwydtwv. Neverthe-
less ThDNT (without saying so explicitly) considers this sense as
peculiar to the LXX, since it does not provide external evidence. Thus,
the most frequent meaning of dtkalwya in the LXX was presumably not
coined by the translators, but was already a natural part of their
lexical inventory.
In this fourth sense, 5tkalwya is generally in plural since the Hebrew
Bible usually speaks about the ordinances of God in the plural. The
word is almost always used in connection with God, so that it mainly
refers to a divine decree. There is nothing intrinsic in the meaning of
S.kalwia making it a divine decree, but the word is used in that way in
the Hebrew and Greek Bible. Stkalwua is usually given by God to men,
or by Moses on his behalf.
Exod 21:1 kal Tatta Ta Stkatwpata a TapabyoeEts EvwTLov
aQuTuV
And these are the ordinances which you shall set
before them.
‘GREEK WORDS AND HEBREW MEANINGS’ 113
A &tkatwpa can be accomplished, kept or fulfilled
Lev 25:18 Kal TOLNOETE TaVTa Td SLKaLwMpPaTd [LOU
And you shall do all my ordinances ...
A dikalwpua canbe listened to, taught, and explained, the latter
especially in the Wisdom psalms:
Ps 118(119):12 8{8aEdév pe Ta Stkatwpatd cou
Teach me Your ordinances.
Most instances of &tkalwua refer to the ordinances of the Israelites, but
some to those of the gentiles.
2 Kgs 17:8 Kal €TopevOnoav Tots Stkalwpaotv Tov €bvav
And they walked in the ordinances of the nations.
The word is also used for the ordinances of the king (1 Sam 8:9,11 etc.;
Mic 6:16).
As one of the stereotyped renderings of vown, 5tkalwya was also used
Hebraistically for meanings of b5vn which are not covered by 8ikalwyua:
[5] = vbw custom*(?), manner*(?)
1 Sam 27:11 Ta5€ Aautd tovet. Kal T48€ TO Stkalwya avtod mdoas Tas
Tye pas As ExdOynTo Aaurd év aypuj tuaiv ddAAoPbAwV
... these things David does. And this is (was) his p5wn
(manner) all the days which David dwelled in the
land of the Philistines.
The translator of 1 Samuel atomistically represented vDwn with
Sikalwpa, even though in this verse the Hebrew had a meaning differ-
ent from its main sense. The best way of representing this Greek word is
to take it as a symbol for the Hebrew v5wn, here used as ‘custom’,
‘manner.’ For a similar usage of étkalwya, see Exod 21:9,31.
Exod 21:9 Kata TO dtkalwua Tov OuyaTépwv trorfjoet abty
According to the v\un (= manner) of the girls he shall
deal with her.
This rendering Hebraistically represents v5wn, cf. the first verse in the
chapter (21:1) kal Tatta Td Stkatodpata d Tapabyoets EvaitLov avToiv.
This meaning is therefore recorded as the Hebraistic [5.] = b»wn custom*
(?), manner* (?). On the other hand, ThDNT records Exod 21:9 as ‘the
law of daughters,’ and likewise the next example (1 Sam 2:13) as ‘the
law of the priest.’
114 CHAPTER EIGHT
[6] = b>wn rightful due(*)
1 Sam 2:13 Kal TO Stkalwya Tot Lepéws rapa Tod Aaod
.. and the vown (rightful due) of the priest from the
people.
Of the known meanings of dikalwua, ‘decree’ or ‘law’ is hardly
appropriate, since no decree or Jaw is mentioned. This sense is therefore
recorded as ‘rightful due’ since in this context the priest receives
something from the people. Once again it is claimed that b5wn is meant,
a meaning which we would record as ‘rightful due.’ Cf. kplots in Deut
18:3 for a similar development: kal aity 4 kplots Tov lepéwv, Ta Tapa
Tov Aaod ... Kal Swoet To Lepet ...
In this entry, a note should probably be added on:
Hos 13:1 Kata Tov A\éyov Edpaty S&ikaLtwpata altos édaBev év TH
[opand kal €8eTo atta TH Baad kal dtéBavev
According to the word of Efraim he took &ikatpata in
Israel and placed them for Baal and he died.
Sikatdpata reflects nnn of MT, possibly read differently (Schleusner:
mint). It is hard to determine the meaning of 6&tkatwya in this verse.
Sample entry for a LXX lexicon
* The asterisk indicates meanings which based on the present knowled-
ge are evidenced for the LXX only. The sequence reflects the logical
order of the presumed development of the meanings. Meanings starting
with a Hebrew word presumably developed Hebraistically by expand-
ing the semantic content of the Greek word.
dikalwpa
1. legal right, justification (2 Sam 19:29); likewise: legal case, cause (2 Chr 6:35 [of God]
Kal tottjoets TO Sikalwya abtav; Jer 18:19 elodxovgov THs muwvis Tod Stkatupatds pou, cf.
also 11:20).
2. justice (1 Kgs 3:28 tod rovetv Bikaluwya; TplBuv Sikarwpatos // dd0t¢ Sikatoovvns
Prov 8:20); cf. modern Greek = ‘justice.’
3. <=2a?> in plural: righteous deeds (Bar 2:19 Td Stkatmpata Tuy twaTépwy abTuy).
4, decree(*), ordinance(*), regulation(*), sometimes in sg. (Num 15:16; 27:11, etc.), but
usually in plural (Gen 26:5, Exod 15:25, etc.), almost always divine (Lev 25:18, Num 36:13,
etc.), given by God (tl@njt [Exod 15:25], €vrédAowat [Num 30:17}), and kept by men (trotéw
[Lev 25:18, Deut 6:24], puAdcow [Deut 6:2, 28:45 etc.]) used in parallelism with évtoAy,
TpdoTtaypya, vouos and vdytov; usually of the Israelites (2 Kgs 17:19), but also of the gentiles
(8: kal é¢vopevOnaav tots Sikardpao Tay ébvav [also 1 Macc 1:13]); also used for the
ordinances of the king (1 Sam 8:9, 11 etc.).
“GREEK WORDS AND HEBREW MEANINGS’ 115
2: = vayn custom” (?), manner” (?) 1 Sam 27:11, possibly also Exod 21:9 kata Td &tkalwpa
Taw BvyaTé pwr trotyoet aut; 21:31.
6. = vewn (rightful) due(*) (1 Sam 2:13 Kal Td dtkalwya Tod Lepéws tapd Tod daod; cf.
kplois in Deut 18:3 for a similar development); cf. Modem Greek ‘dues’, ‘fees,’
2. The dpordoyéw group
Due to their Hebrew background, most of the words deriving from the
stem Opodoy- in the LXX carry two different meanings, carried through
into the New Testament and patristic literature.
In some way or other the three basic meanings ‘to acknowledge/
confess, promise, give thanks’ are recorded in the various lexica. The
sense ‘to give thanks’ is recorded either as a separate sense or as a
subsense of ‘to acknowledge,’ viz., ‘to acknowledge thanks.’
LS]
dpodoyéw (Act.)
I agree with, say the same thing as
I] 1. correspond, agree with
2. agree to, grant, concede, avow (gratitude), acknowledge, confess
3. agree, promise to (do)
dpodoyla
1. agreement
assent, admission, concession
agreement, compact
vow [LXX Je 51(44).25 (pl.)]
conformity with nature
oh So
OuUod\dyus
1. agreeable to, in unison with
2. confessedly, openly, LXX Ho 14.5; also €E boddyou Pb. 3.91.10
éfopodoyéopa
1. confess, admit, acknowledge
2. make grateful acknowledgements, give thanks, sing praises (LXX, Philo, NT)
éFouoAd ynots
admission, confession, confession of gratitude Philo
dvGopoAoyéopat
I make a mutual agreement
II 1. confess freely and openly
2. admit, signify
3. assent, agree
4. return thanks (to God) LXX, NT, xdptv dvé. Plut. Aem. 11
1.7. Pring, The Oxford Dictionary of Modern Greek (Oxford 1965).
116 CHAPTER EIGHT
dvOopoAd ynois
1. mutual agreement
2. confession, admission, testimony
BAGD
dporoyéw
1. promise, assure
agree, admit
confess
declare (publicly), acknowledge, confess
praise w. dat. Heb 13.15
me Ww Nh
dpodroyla
1. act. confessing (as an action)
2. pass. confession, acknowledgement (that one makes)
éfopodoyéw
1. act. promise, consent
2. mid. a. confess, admit
b. acknowledge
c. praise (deriving from mng. a)
MM
dpordoyéw
agree with, acknowledge (+ parallels). ‘... With the usage “praise”, “celebrate”, in Heb 13.15,
a a”
which Grimm-Thayer pronounces as “unknown to Grk. writ.,” we may compare the
somewhat similar phrase d,0d0ya yd pita in petitions, e.g. P. Ryl. II. 11432 (c. A.D. 280)
Sts ... MY TUXQ gov xdpitas dpodoyeiv Surynda, “that I may avow my gratitude to your
fortune”, P.Oxy. 1.6722 (A.D. 338 ... xapttas dpodroyfouwpev, and the Christian letter P.Oxy.
1.9398 (iv A.D.) ... tds xdpitas dpoldoyodrtals ...’; promise (+ parallels).
dpodroyla
contract, agreement (+ parallels)
éFoporoyéw
acknowledge, avow openly, consent, agree (+ parallels); ‘In the LXX the idea of “give thanks”,
“praise”, is prominent: cf. in the NT Mt 11.25, Lk 10.21, and perhaps Phil 2.11 (see
Lightfoot ad loc.).’
dvOoporkoyéopat
agree, answer, acknowledge (+ parallels)
One of the main questions relating to the various usages of this word
group is the background of the meaning recorded as ‘to give thanks.’ The
distinctions between the various senses are admittedly very subtle. The
verbs ‘to confess’ and ‘to give thanks’ are two separate entities;
however in some way the two groups are possibly related. Is there a
middle path in the Greek language which may be defined as ‘to
acknowledge thanks’ and which would explain the co-existence of
"GREEK WORDS AND HEBREW MEANINGS’ ees
these two senses? If so, one would be able to see how one sense developed
from the other.
The meanings ‘to agree, concede, acknowledge, confess, promise’ are
found in all of the Greek literature, including the LXX and the
literature based on it. Some examples follow referring to the simplex
and its composita, both nouns and verbs:
Wisd 18:13 ETL TH THY TOWTOTOKWY dAEBpW WYOADyTGAV PEot vLOV
adv elvat (admit)
Sir 4:26 UN) aloxuvOris duodoyrjoat €d’ Auapttats cou (confess)
Jer 44(51):25 = mtototicat Totjoopev Tas duodoylas hudv ds
wWpLodoyoapev (promise)
Dan 9:20 ... Kal EEoodoyotptevos Tas aGuaptlas pou (confess)
On the other hand, the meaning ‘to give thanks’ is found only in the
LXX and literature based on it, signifying that it was probably created
in the Hebrew realm:
1 Esdr 5:58 Kal €duvnaay 6.’ buvwv duodroyobvtes TH Kupli, STL ...
(= 1797, give thanks)
1 Esdr 9:8 S6Te doroytav S6Eav tu kuplw (‘thanks’)
Ps 6:6 ev 6€ Tu G5y ths EEooAOyHoeTal oor (= 7717)
Ps 9:2 €€ouodoyrjooual coor Ktpre év Sdn KapSla pov Sunyrjoopat
TdvtTa Ta Bavudoid cov (= 17117)
Dan 2:23 GOL, KUPLE TUV TaTé pw pov, €EooAoyoupat Kal alva
Naw) NT NAIR APR 7?
1 Chr 25:3 év KLvUpG dvaKpoudéuEvoL €EoLoAGynotv Kal alveoiy Tu
Kkuptw (= 1717)
Ps 78(79):13 = dv8opodoynodpe@d aot els Tov aldva
It is not easy to decide that a word or word group has been created by
the LXX translators. Such decisions are based on negative evidence
relating to the extant Greek literature, although new evidence may
always turn up in a previously unknown papyrus or inscription.
In light of the aforementioned examples it seems to us that the
special use of the 6uod\0yéw group in the LXX as ‘to give thanks’ is
sufficiently evidenced, both for nouns and verbs, the simplex and its two
composita. It should however be added that in the lexica this use is
also recorded incorrectly as ‘to praise.’ The basic meaning ‘to give
thanks’ should be adhered to, and if the word is sometimes used in
parallelism with ‘to praise,’ the juxtaposition does not create a
meaning ‘to praise’ for Oypodoyéw. This criticism applies to the
dictionaries as well as to a study by J. Ledogar, who recorded éEo1od0-
118 CHAPTER EIGHT
yéouat as ‘to praise,’ discussing this word together with the other
words for praise in the LXX.@
The sense ‘to give thanks’ also occurs in the New Testament, e.g.:
Mt 11:25 ... 6 "Inoous eltrev, EEopodoyotpyal oor, taTEp KUpLE TOU
ovpavod kal THis yrs = Lk 10:21
This special meaning is not evidenced outside the realm of the LXX
which includes the New Testament and other literary sources
dependent upon it. Some scholars, however, claim that the papyri
contain an intermediary stage between the meanings ‘to confess’ and ‘to
thank,’ viz. xdpttas or xdptv dpodoyetv, translated as ‘to acknowledge
thanks.’ Such a phrase occurs, for example, in:
P. Ryl. 1] 114,32 (c. CE 280) &t1ws ... tT Thy Cou xdpitas dpodoyety Suvnbid
This approach is reflected in MM, quoted above, and likewise in LSJ
which does not providea separate entry for duodoyéw as ‘to give
thanks,’ but includes that meaning in sense II 2 (agree to, grant, concede,
avow [gratitude], acknowledge, confess). For the sense ‘to avow
gratitude’ LSJ quotes the same papyrus as MM. However, xdpttas or
xdptv dpodoyetv means ‘to avow gratitude,’ but that gratitude is
expressed only by the noun, viz. xdpitas or xdptv, and not by the verb
oyodoyetv. Therefore the basic meaning of that verb remains unchanged
by the usage of the papyri, viz., ‘to acknowledge,’ and no intermediary
sense is detected.
Since there are no parallels for the sense ‘to give thanks’ outside the
LXX, we must try to locate the solution for the lexicographical problem
within the biblical realm. It seems that this meaning has developed
from an etymological procedure which identified two Hebrew roots,
1711/%, ‘to confess’ and 771/7- ‘to thank.’ In Hebrew these two senses are
expressed by the same root. These two verbs derived from a common
source, or one sense developed from the other one. To find out the exact
relation between the two senses of 77)/° as ‘to thank’ and ‘to confess’ is a
matter for Hebrew linguists and exegetes, who usually record the two
senses as separate entities. 7717 in the hiph@l is used generally as ‘to
thank,’ but also a few times as ‘to confess,’ and 771na in the hitpael is
more frequently used as ‘to confess’ than ‘to thank.’
The translators did not distinguish between these two different
senses. They derived both of them from the meaning ‘to confess,’ and for
- J. Ledogar, “Verbs of Praise in the LXX Translation of the Hebrew Canon,” Bib 48
(1967) 29-56.
“GREEK WORDS AND HEBREW MEANINGS’ 119
this sense they used 6duo\0yéw and compounds, thus not recognizing a
separate meaning ‘to give thanks.’
The concepts of ‘givings thanks’ and ‘confession’ are somehow
related, but it cannot be claimed that the translators also made this
link on aconceptual level. Therefore, the theological approach
explaining this rendering as ‘a Semitic linking of confessing sin and
praising God,’ as in TRDNT, is not acceptable. Rather, the translators’
decisions must have been based on an etymological procedure involving
even 717, glory, honor.
Ps 95(96):6 €EopoAdynats Kal woardtTyns EvasTrLov abou
yp) 7711717
Honour and majesty are before him (RSV).
Ps 103(104):1 €E€opoddynovw kal evmpétevav évedUow
nwa? WM TN
Thou art clothed with honour and majesty (RSV).
These examples show that the translator of Psalms, who otherwise
was aware of the meaning of 117 (cf. 20:6, 44:5, 144:5), derived the word
in these two verses from 171, and used the usual equivalent for that
word in the LXX.
The Greek translators thus used the d20\0yéw group for various words
and meanings of the 7717 group. It is not impossible that they were
aware of the difficulties involved, since for the meaning of
‘thanksgiving’ the simplex is used much less than é€opodoyéopat and
dvOopodroyé op.at. Possibly these two composita were reserved for these
special meanings.
The translation thus reflects the merging of two meanings in the
wake of the Hebrew in a way which does not suit the Greek language.
This is a true Hebraism which cannot be expressed well in a conven-
tional lexicographical description of the LXX. The only way to express
such a Hebraism would be to describe it as ‘= 1710, to give thanks.’
In our view, a meaning ‘to give thanks’ for €€opodoyéw did not exist
at the time of the translation, as the translators did not express such a
meaning in their rendering. But such a sense was bound to develop. How
else could a later generation explain a verse such as
i Chr 25:3 Ev KLVUpa dvakpovdpEvor E€onoASyyotv Kal alveoty To
Kuplw
In some verses somehow the sense ‘to confess’ can be maintained, but the
context makes such a claim impossible in this verse (as well as in 2 Sam
22:50 = Ps 17 (18):49; Ps 32 (33):2). Thus for the readers of the LXX there
existed a new sense of dpodkoyéw, with its compounds, as ‘to thank,’ and
120 CHAPTER EIGHT
the respective contexts made it easy for the readers to expand that
meaning to the sense ‘praise,’ even if this meaning was not intended by
the translators.
This new understanding of the Greek word has been expressed well in
the anonymous Comm. Tura (P. Colon. Theol. 1) 5:13 on Ps 30(29):5:
éEquoddéynois évtavdda ok fh €Eaydpevots eth duaptripaowy dé yeTat.
MéyeTat pév yap kal atm éEopoddynats ... ds Em TO TAETOTOV obv ev TH
ypady EEonodSynats 1a THs AEEEwWS onpalvetat fh evbxaptotia.
The quotation from Isa 45:23 in Rom 14:11 shows the problems involved:
Isa 45:23 KaT’ €uauTov duvvw ... OTL Eol Kdpwer av ydvu kal
éEonodoyyoetat (SCLC duettar) tdoa yucca To Be0s
(74 Atkatoavun Kal 86Ea pds avtov HEovowv)
WP 2 VIVA 7S: PS YIN) 2 ee “AVION 3
Rom 14:11 Cd eyw, Aéyet KUpLOS, Ste ELol Kaper dv yévu kal tdoa
y\ioaa €EonodoyhoeTat Tu bed (dpa obv Exaotos hav
TreplL EauTov Adyou Suet Tu) BEd)
... and every tongue shall give praise (or confess) to God
(RSV).
Within the context of the LXX the Greek verb means ‘to give thanks,’
cf. the parallel phrase of the bowing of the knees. This is also clear
from the continuation of the verse (24) \éywv Atkatootvn kal &6€a pds
autov TEovotv. In Romans, however, the verse has been taken in a
different sense. The continuation of the quotation (‘so each of us shall
give account of himself to God’) makes it clear that Paul took the Greek
as ‘to confess.’ This doubt regarding the meaning is further continued in
the RSV in Romans where the word inthe quote from Isaiah is
translated as ‘give praise,” but the note refers, more correctly, to
‘confess.’ Here, as elsewhere, the way in which New Testament
quotations understand the words of the LXX cannot be taken as the only
way of interpretation.
SAMPLE ENTRIES
dpodoyéew
admit + inf. Wisd 18:13, 2 Macc 6:6.
confess Est 1:1; &’ Guaptlats aov (Sir 4:26), rhv ém6uplav atta (Sus 14 6’; cf. LXX éE-)
promise (or vow” [cf. d40d0yla Lev 22:18]) Jer 44(51):25.
= 110 (?) give thanks* + dat. (to) oof (I Esdr 4:60), tT kuplw (5:58), abs. (Job 40:14).
dpodoyla, €&-, EEopoddynois, av, dvboyoAdynots ; bibl. R.J. Ledogar, “Verbs of Praise in
the LXX Translation of the Hebrew Canon,” Bib 48 (1967) 29-56.
J Peper
“GREEK WORDS AND HEBREW MEANINGS’ Pal
dporoyla
1. promise (or vow* [= 771]) Lev 22:18, Jer 44(51):25.
2. thanks, &te duodoylav &dEav to kuplw (1 Esdr 9:8).
3. = 127) voluntary offer* pépw (Deut 12:6B), ddyw (12:17), Toréw (Ezek 46:12) (cf.
duod6yws = 7271 Hos 14:5); for a similar semantic development, cf. Td Exovotov = 7273
voluntary offer*.
Gord yws
voluntarily ayamiow avtots dod6yus (Hos 14:5)
éfoporoyéopiat
1. confess Tag duaptlas pov (Dan 9:4, 20) thy d8dvnv abvtod (Sus 14; cf. 6’ simplex).
2. =n7 (?) give thanks* (often with musical instruments, Ps 32:2) usually + dat. (to),
frequently in Psalms, + oot (6:6), + kuplw (9:1), + Ta dvdépaTl cov (53:6); + acc. Ta Bavpdod
cov Ps 88:5; + €vavte xuplovu (2 Chr 7:6), €vavttov (Dan 6:10[11]); abs. (2 Chr 31:2); + 6tT1
(Dan 3:90); in 1-2 Chronicles often paired with alvéw.
— simplex
éfouordynas
1. =a thanksgiving,* thanks,* &8wyur (Josh 7:19), // afveots (1 Chr 25:3), padyos els
éEopoddyno.v (Ps 99[100]:1), all based on the etymological understanding reflected in
EEopodoyéopat 17%1—esp. visible in next usage; 1 = (‘glory’, but derived from nm, as above)
thanks”, thanksgiving* €€opoddynots Kal pardms évatov abtod (Ps 95[96]:6 thanksgiving
and beauty*, before him), €E€opoddynoiv kal evmpéterav Evedvow (103:1 ‘you have clothed
yourself with thanksgiving and glory’), 148:13.
dvOopodoyéopat
1. confess | Esdr 8:88, Sir 20:3
2. =a (1) give thanks™ + dat. (Ps 78[79]:13, Dan 4:34); + els (3 Macc 6:33; étl TLL).
dvGoporS ynois
1. = 700 thanksgiving™* 2 Esdr 3:11, 6(8ut (Sir 17:27).
3. Empavns
Neither LSJ nor MM refer to any special use of this word in the LXX.
LSJ
I 1. coming to light, coming suddenly into view, appearing
2. of places and things: in full view, visible
3. manifest, evident
IY 1. conspicuous, notable, distinguished, famous
2. remarkable
3. a title of divinities
BAGD
splendid, glorious
22 CHAPTER EIGHT
From the outset there appears to be no special LXX meaning for
émigdavys, whose general meanings are ‘appearing,’ ‘visible,’ ‘evident,’
‘famous,’ and ‘remarkable.’ The meaning ‘evident,’ ‘visible’ appears to
be well established in Prov 25:14:
WoTrEp dvEepLoL Kal vedy Kal veTol Eripavéotator
... like winds and clouds and clearly visible rains
This use is close to the etymology of étmidavis, being derived from
étidatvw. In other places in the LXX the word is used as ‘glorious,’
another basic meaning of the Greek word, derived from its primary
sense. Thus the temple is ‘glorious’ in 2 Macc 14:33:
... Kal Lepov €vtatéa tu Atoviow étidavés dvaotiow
KUpLos is glorious in 2-3 Maccabees (e.g. 2 Macc 15:34, 3 Macc 5:35).
Also the ‘day of the Lord’ in Joel 2:11 (cf. also 3:4) is seemingly
described as ‘glorious’ (but see below):
S.6Te LE yaAy 1 Hepa Tot kuplou, weydaAn Kal €tLdavys
apdddpa
Likewise, the ‘name’ or ‘fame’ of the people of Israel in 1 Chr 17:21 is
seemingly called ‘glorious’:
... 0 8€0s ... Tot O€c8at EauTa dvoua péya Kal Etipavés
This applies also to the dpaots, that is, the ‘appearance,’ of the angel
in Judg 13:6, both in manuscripts A and B:
Judg 13:6A dvOpwtros Tot Beot AAGev mpd LE Kal 4 Spaats abtod us
Spaois ayyédou Tod Seot Etidarviys ofdhpa
Judg 13:6B dvOputros beot AAGev Tpds pe Kal El80s adTod we ElSos
ayyéXov Geo doBe pov apdSpa
The different contexts indeed allow for an interpretation of the
aforementioned instances of €tmidavijs in the LXX as either ‘glorious,’
‘evident,’ or ‘famous’ and the word should thus be recorded according to
its natural meanings.
However, an analysis of Hab 1:7 shows that at least in this case the
Hebrew should be consulted as well, and the other verses reassessed
accordingly.
Hab 1:7 (6 TO EBVOS TO TLKPOV ... TO TOPE UdULEVOV ETL Ta TAGTH THS
YS TOU KATAKANPOVOLL OAL OKHVULAaTAa OvK abToD)
poBEepds Kal ETILavys EGTLV
MT NT NT) ON
“GREEK WORDS AND HEBREW MEANINGS’ 123
In his description of the coming of the Chaldeans in 1:5-10 (117),
Habakkuk has little good to say about this people. They kill and
destroy, they mock at the kings and there is ‘no end’ to all of this. The
Chaldean people are described in the LXX of v. 7 as poBepds kat
étidavris. It is understandable that the Chaldeans should be called
goBepds, that is, ‘frightening,’ ‘terrible,’ but what does the next word,
émLdavys, mean in this context? Are the people ‘conspicuous,’ ‘evident,’
or ‘famous’? Or should we rather take €t.gavrjs as the opposite of its
main meaning, that is, ‘infamous’?
When analyzing the linguistic problem, the Hebrew text should be
consulted as well. The translator understood his text incorrectly, and
that misunderstanding gave rise to the present translation. The MT of
Habakkuk described the Chaldeans as a frightful and terrible people:
NT NT) ax, but the translator derived x11 from the root 77, to see, and
not from x, to fear. It does not really matter if we call this etymology
‘wrong’ or ‘idiosyncratic’; it differs from the usual understanding of
that word, and the confusion of these roots is known elsewhere in the
textual and exegetical tradition of the Bible. Thus, instead of
describing the people as ‘terrifying,’ the Greek translator described
them as ‘visible.’ Reference to the Hebrew is necessary in order to
understand the background of the translation; otherwise the Greek
lexicographer may ascribe to the word an unusual meaning, including
‘infamous.’ The mistakes of the translator are not of primary concern to
the lexicographer of the LXX, nor does it matter to him if the Greek
text makes sense as long as the words themselves are understandable.
However, in this case one needs to know what the translator had in
mind before determining the meaning of this éttdavys. It may not be
very appropriate to say of the fearsome people of the Chaldeans that
they are ‘conspicuous,’ but this is how the translator took x71. Thus
étmipavt|s should be taken as meaning basically ‘conspicuous’ regardless
of whether or not such a meaning makes sense in the context (see Tov,
“Dimensions”*). Accordingly, a LXX lexicon should nevertheless record
this €mipavis as another instance of ‘evident’ or ‘conspicuous.’
In Zeph 3:1 one is less certain about the meaning of the Greek word:
70 4 ettthavys Kal dtrok€eAUTPWLLEVN, T] TOALS T) TEPLOTEPA
Alas the conspicuous (?) and redeemed city, the dove ...
MPT TYO TID ATV I
Woe to her that is rebellious and defiled, the oppressing
city (RSV).
124 CHAPTER EIGHT
The context is that of a city in despair. It did not listen to the Lord, its
leaders are corrupt and there is little good to say about that city. How
should étig@aris be taken in this context? Once again, one thinks at first
about the city being ‘glorious.’ Obviously this is not very appropriate in
the context, but the Greek word could still be taken as such if it be
understood ironically. However, the Greek context warns one against
ascribing exegetical inclination to the translator. The verb 2x1, used in
MT as ‘defiled,’ is taken by the translator as derived from another
meaning of that root, viz., ‘redeemed.’ And 7177 Yyi, ‘the oppressing
city,’ has been misunderstood as ‘the city, the dove.’ This phrase makes
as little sense in English as it does in Greek. Therefore, because in some
details the translation makes little sense, from the outset it is not
likely that €migavis reflects any special understanding of the context.
As in the preceding example, the translator derived 7x71 from the root
m7, rendering it as if the city were ‘visible.’ We should therefore take
éTLmavis as ‘conspicuous,’ ‘visible,’ and not as ‘famous’ because the
translator did not have that meaning in mind.
It seems that also in other places translators derived x71 from the
root 7X7 as if reading 7x71, translating the word as étidavys. In fact,
this also applies to the examples which at first sight were considered
uncomplicated. Thus the ‘day of the Lord’ is named étdari¢ (Joel 2:11;
3:4), ‘conspicuous’ and not ‘famous.’
Sample entry for a LXX lexicon
1. evident dveyor kal védn kal beTol (Prov 25:14), conspicuous Td €6vos (Hab 1:6), wd
(Zeph 3:1); hwépa Tod kuplou (Joel 2:11, 3:4); dvoya (1 Chr 17:21), Spaois Judg 13:6).
2. glorious lepdv (2 Macc 14:33), kuptos (15:43, 3 Macc 5:35).
4. dpOplfw
dp8pl{w does not create any specific difficulties in the Greek language
in which it occurs rarely. Only in the literature which is somehow
related to Hebrew sources do some problems arise.
Two different forms are used in Greek, dp6petw, the Attic form, and
dp6p(Cw used in the LXX and New Testament, as well as in the
literature dependent on these sources and, in a different form, as
opOpl{w and dpt{¢w in an ostracon and a papyrus. The entry in LSJ gives
little information, but LSJ, Supplement provides a detailed entry with
three different meanings.
‘GREEK WORDS AND HEBREW MEANINGS’ 125
LS]
= 6pOpevw, LXX Ev. Luc. 21.38
dpbpetw, (Sp8pos) lie awake before dawn, Theocritus, Eur.; dp6pevecbar kadovorv ol
"ATTLKOL TO AUXYW TPCOKETobaL, Mply hépav yevéobat, Phryn. PS p. 93 B
LSJ, Supplement
la (= Attic dp8pevw, Moer. p. 272 P) rise up early, LXX Ge 19.2, Ps 126(127).2
metaph., be eager or earnest, dp8plCwv Je 25.3
2 go early, els tétrov ib. Ge 19.27; go eagerly or earnestly, tpd¢ tiva ib. Ps 62(63).2, Si 39.5,
Ev.Luc. 21.38
3 Tporyev opdplCwy kal dy(Cwy at morning and evening Thd 1 Ki 17.16
BAGD
(Moeris p. 272 dp6pever "ATTriKcic, dpBplCer "EAAnviKus) impf. dpOpiCov be up or get up very
early in the morning ...6 kabs wp8piCev mpds abtdv év To Lepul dkovetv altod the people used to
get up very early in the morning (to come) to him in the temple and hear him Lk 21.38 (dp. pds
Tiva also means gener. seek someone diligently: Job 8.5; Ps 77.34; Si 4.12; Wi 6.14; Test Jos 3.6)
MM
... According to Thumb (Hellen., p. 123) the dependence of the verb on the Heb. o-avn in
the sense of ‘rise early’ is very improbable, and reference is made to such analogous verbs in
M Gr as vuxtopevw ‘work throughout the night,’ and peanpepidfw ‘do something at
midday.’
The meaning ‘to get up early in the morning’ is the basic sense of this
verb and as such it occurs also in the LXX. As a rule it reflects ow in the
Hiph4l, usually in the form o3w7, that is ‘and he got up early in the
morning,’ often linked pleonastically with 1722. For example,
Judg 21:4 AB kat €yéveto (A: év) TH Etravpiov Kal dp8p.cev 6 Aads
On Ww Dwr) NNN 17)
Gen 19:27 wspOprcev S€ ABpaap Td Trpwt
9P22 O7728 D3v
Exod 34:4 Kal dp8plaas Muvaijs avéBn els 1d Spos TO Ziva
70 TA OX Dy PAD nw oDws
See further:
Ps 63(62):2 ‘O Beds 6 Beds Lov, TPds oe SpbpiCw
God, my God, for thee I get up early in the morning
ADNWN INN 0PX OPN
O God, thou art my God, I seek thee (RSV).
Isa 26:9 Ex vuKTos Opbpllet TO TYE DG Lou Tpds aé, 6 Beds
FTOWN PRN 7A rN AN 7772
My soul yearns for thee in the night (RSV).
Hos 5:15 év OACWet avTuv dpBpLotcat pds Le AE-yovTEs
126 CHAPTER EIGHT
"nw? oO? AN?
And in their distress they seek me (RSV).
In the last three examples, MT uses the verb 1nw, ‘to seek,’ but the
Greek translators derived these forms from the noun 1n¥, ‘morning.’ The
resulting translation, referring, like 1nw, to the ‘morning,’ is quite
possible. Thus, instead of the verb of the Hebrew text which should be
taken as ‘to seek earnestly, diligently,’ the LXX uses here dp0p(Cw,
which should probably be taken as ‘to get up early in the morning’ for
someone. These two verbs are quite different, so that we read, e.g., in
the LXX of Ps 62:2 ‘My God, My God, early in the morning I get up for
thee’ instead of MT ‘O God, thou art my God, I seek thee.’ The Greek
verb is used in a slightly different way from its use in classical Greek,
namely not as just ‘to get up early,’ but as ‘to get up early for someone.’
This particular use is created by the etymological rendering of the verb
nw from nw, ‘morning.’ Accordingly, there is no need to ascribe to the
Greek verb a new meaning such as in BAGD (‘seek someone diligently’)
or in New Documents, vol. 1 (Sydney 1981) 86 (‘to come’). These sources
adapt the meaning of the Greek verb to the context; BAGD ascribes to
dpOplfw the meaning of its Hebrew counterpart, a procedure which is
quite questionable from a methodological point of view.
This use of 6p8p{€w with a preposition is possible in Greek, as is
shown by the text of the ‘Amsterdam ostracon,’ discussed at some length
in New Documents, vol. 1, p. 86: (va ulvys abtév, él yap dpTi¢er mpds oe
auptov (22.7-8 [II]). In that ostracon there appears to be a misspelling
(omission of the p), also known from the manuscripts of the LXX (in Ps
126:2 Codex S has dp6{w), but nevertheless it is clear that dp8p{(w is
meant. I do not think that in any of these places the original meaning of
the Greek, that is ‘to go early’ or ‘to rise early’ has been lost, and in any
event it would be hard to prove such an assumption. A comparison with
these documents, then, shows that in the use of the verb with a
preposition no Hebraism is involved since a similar construction is found
in two external sources not dependent on the LXX (P. Amst. 22. 7-8 [ii CE]
and P. Mil. Vogl. II 50.13 (I) topedou obv dpOlaas els [so rising up early,
go to ...]).
The verb occurs also once in Luke: 6 Aads dpOptoev mpds abtov év Ta
Lepui dkovetv abtot (21:38), which should be translated as ‘The people
rose up early in the morning to (see) him in the temple,’ or, as in the
RSV, ‘And early in the morning all the people came to him in the
temple.’ The use of the verb in Luke follows the usage of the LXX, but
since this usage is also found outside the LXX, Luke’s dependence on the
LXX cannot be proven regarding this detail.
“GREEK WORDS AND HEBREW MEANINGS’ 127
Sample entry for a LXX lexicon
1. get up early in the morning (cf. 6p8pos) Judg 21:4, 1 Kgs 5:3, Tob 9:6, usually
pleonastically joined with 76 towl (following MT 7721 o72w7) Gen 19:27, 20:8, Exod 24:4;
often the participle is used together with other verbs dp8ploas 5¢ Muwvarjs 76 Trpwi
wKxoddunoev Bvotaotrpiov (Exod 24:4), kal dpSploas 'I noots To tpwl éteokéwato Tov Aaév
Josh 8:10); also with verbs of motion kal dp8plaag Muvors dvéBn els Td Spos TO Diva (Exod
34:4), kal dpOplaavtes Td Trpwl avéBnoav els Thy Kopudhy Tot Spoug (Num 14:40); abs. with
els get up early in the morning (and go) to Sp8picev S¢ ABpaay Td trpwl els tov TéTroOv (Gen
19:27), dpAp.ette abptov elg Thy d8dv bay (JJudg 19:9), with év: kal dpOploate év 1H 685 (1
Kgs 29:10); tpds tiva get up early for someone (God)'O 6eds 6 beds Lov, 1Tpds at dpOpl(w (Ps
62[63]:2), €k vukTds dpBplCet TO TVEDUA Lov Tpds o€, 6 Beds (Isa 26:9), év OAL WEL avTiy
dpBprovat pds YE Aéyovtes (Hos 5:15) (in all these cases, as well as in Job 8:5, Ps 77[78]:34,
an aspect of yearning is extant in the Hebrew verb [nnw, to seek], but not in the Greek verb
where this aspect is expressed by the preposition). Cf. O. Amst. 22.7-8(I) {va lungs abrév,
éml yap dptlCet mpd¢s ce aliptov and the discussion in New Documents 1, p. 86.
The third sense listed in LSJ (4p8p(Qwv = ‘at morning’) is not distinct
from the others and should probably be listed together with them.
But there is one other sense (‘early’, ‘earnestly’) listed as 1.b in LSJ
and as a separate meaning 2 in the sample entry which should be taken
as a Hebraism.
2 Ch 36:15 kal €€atéotetrev KUpLOsS ... dpOplCwv Kal dTOoTEAWY
Tous dyyédous avTot
And the Lord sent ... his messengers dp8pl(wv and
sending.
mw) Ow PIRI TI... AYP Nw
Jer 25:3 €ddAnoa (sc. kKbptos) Tpds Luds dSpOplCwv kal Aéyur
I (scul., the Lord) spoke to you rising up (?) and saying.
7271) DWN OD°7X VIAN
ibid. 4 Kal dtéOTEANOV TIPOS LUGS ToUs SoUAOUS [Lou TOUS
Tpogitas Op8pov atoaTé \Awv
nw) OWA OXI yay 2D nx ODN A Ww
This dp8p{Cwv has been recorded as ‘o3wi = early, earnestly*’ with the
following implication: The construction in which dp8plCwv occurs differs
from that of the other instances, and is used Hebraistically. In 2 Chr
36:15 it cannot be taken as ‘to get up early’ as that use would not fit the
subject, the Lord. Moreover, the participle dtooté\\wv is awkward as
it is identical to the main verb. This applies also to the verse in
Jeremiah. From the continuation of the verse (v. 4) we understand how
dp8plCwv is to be taken. It is apparently used as if it were an adverb, just
as in the Hebrew, and parallel to v. 4, where the Greek text renders the
128 CHAPTER FIGHT
exactly same phrase n)?v1 awn with an adverb dp0pou dtrooTéAwy,
sending early in the morning.
In these two verses dp8piCwv is thus used Hebraistically as a symbol
for D2W7.
The second part of the sample entry reads:
2. dp8pl{wv = orwn early, earnestly*, of the sending of God’s message to mankind, used
ungrammatically as a participle (// 5p8pou used in the same situations) kal ¢Eaméate.hev
KUPLOS ... OpOplCwv kal dtroatéAAwv Tabs dyyédous avTov (2 Chr 36:15), €ddAnoa (sc.
KUpLOS) Tpds buds dpOplCuv kal Aéywv (Jer 25:3, cf. v. 4: kal dtéoteddov pds buds Tous
Sob\ous jLov TObs Tpodt|Tas SpBpov dtrooTéAAwV).
Ill. TRANSLATION TECHNIQUE AND EXEGESIS
CHAPTER NINE
COMPOUND WORDS IN THE SEPTUAGINT REPRESENTING TWO
OR MORE HEBREW WORDS
Greek, like other Indo-European languages, has several kinds of
compound words (CWs). The majority of these CWs start with a
preposition (preverb); others are composed of combinations of an
adverb /adjective/noun and adjective/noun/verb.! The LXX translation
also contains CWs, and as biblical Hebrew does not have such
composites, the relation between Greek CWs and their Hebrew
equivalents is analyzed here.
CWs in the LXX represent one, two, or, in rare cases, more Hebrew
words, e.g., ypap.atoetoaywyevc - Vow (Deut 1:15, etc.), TTEpo-Pvew -
Jax Dy (Isa 40:31) and dv-uto-véntoc - 17 %y 22 (Sir 11:5)—a hyphen
written between two elements of a CW distinguishes between elements
representing separate Hebrew words. The group of CWs representing
one Hebrew word is the largest, and a study of such CWs is in order
within the framework of a study on the choice of equivalents. The
present study is concerned mainly with CWs representing two or three
Hebrew words. Its primary purpose is to present the evidence (below,
section 7) to be introduced by an analysis of the use of CWs in the LXX,
their distribution in the various books, the translators’ approaches to
them and the coining of new words.
Before embarking on an analysis of the phenomena, the nature of
this study and the definition of the CWs are first clarified. The point
! The evidence pertaining to Greek compounds is available both through lexica and
reverse indexes such as C.D. Buck and W. Petersen, A Reverse Index of Greek Nouns and
Adjectives (Chicago 1944), which lists CWs according to their final constituents. Evidence
for the LXX is available through the concordances (see n. 16) and X. Jacques, List of
Septuagint Words Sharing Common Elements (Subsidia Biblica 1; Rome 1972). For linguistic
studies of CWs, see H. Jacobi, Compositum und Nebensatz (Bonn 1897); A. Debrunner,
Griechische Wortbildungslehre (Heidelberg 1917); E. Schwyzer, Griechische Grammatik ]
(Miinchen 1953) 425 ff. (with bibliography).
This situation does not apply to modern Hebrew, cf. H. Rosen, Comptes-rendus du
GLECS 10 (1965) 126-135. For biblical Hebrew see, 1.a., Gesenius—Kautzsch, Grammar, § 81d,
152a and the literature quoted there.
132 CHAPTER NINE
of departure for the present study is the LXX rather than MT because
the grammatical category described exists in Greek and not in biblical
Hebrew and because the relationship between the employment and non-
employment of CWs in the LXX cannot be described statistically (the
same applies to the description of, e.g., the syntax of cases or genitivus
absolutus in the LXX). The study is limited to the CWs in the LXX and
therefore it does not compare the form, frequency, structure, and
meaning of CWs in the LXX with CWs elsewhere. The term ‘compound
word’ can thus be employed in an unorthodox way. For example, akpé-
8ujLog is a true compound consisting of pakpdg and 8ujd6c; on the other
hand, the similar pakpo-Oupéw may look like a compound but is in fact a
derivative of paxpdé-8uynoc. For the translators, however, no such
distinction existed between words of the paxpdé-8upoe type and that of
Laxpo-Oujew. Both words were used to represent combinations of two
Hebrew words, see 0°58 778 - LaKpd-8ujLog in Exod 34:6 and passim and
DX 77Xn - wakpo-OSupéw in Prov 19:11. Therefore in this study, words of
the latter type are also treated as CWs.
1. CWs in the LXX
The LXX contains many CWs which represent single Hebrew words.
Often one Hebrew verb is rendered by a composite verb (verb + preverb)
because it represents the meaning of the Hebrew in the best possible
way. Thus ‘to descend’ is represented by kataBafvw, ‘to ascend’ by
dvaBafvw and ‘to embrace’ by teptAauBdvw. In other instances the
preverb expresses a special meaning or nuance of the Hebrew word.?
Thus while ow is generally rendered by t67pt in Ruth 3:3, repiTlOnpt is
used for the putting on of clothes. Sometimes x7? is rendered by
Tpookadéopal (e.g., Gen 28:1), while its usual equivalent is the simplex
kadéw. In yet other instances the CW expresses a composite idea, e.g., a
d\oKaUTwWLG is ‘an offer burnt in full’ (= 171v) and @nptdAwrtoc is
‘something caught by a wild animal’ (= 757).
Generally when translators rendered one Hebrew word by a CW,
they did so either to express a composite idea or to conform to the norms
of the Greek language. However, when two or three Hebrew words were
rendered by a CW, the motivation of the translator was probably
different. In the latter case, many translators probably thought that
certain combinations were best represented by one composite word, even
though a two-word representation would not have been unidiomatic.
3 CE. especially M.L. Margolis, “The Greek Preverb and Its Hebrew-Aramaic Equiv-
alent,” AJSL 26 (1909-10) 33-61 and below, section 7 II.
COMPOUND WORDS 133
Sometimes there was also an intrinsic reason for the use of a CW:
although biblical Hebrew possesses no CWs of the same type as the
Indo-European ones, many word-pairs behave as CWs on a syntactical
level. This applies in particular to word-pairs whose first constituent
is either a word in the construct state such as 787 Nd* - €v-TrpdawTl0G
(Gen 12:11) or a participle such as ywd 34x - ¢-A-ayapTHwy (Prov 17:19).
Two factors were instrumental in the employment of CWs. On the one
hand, the CW might have easily suggested itself as an equivalent for a
combination of two (three) Hebrew words. On the other hand, its
employment might have resulted from the translator’s stylistic
inclination to search for a compound Greek word as an equivalent for a
closely related Hebrew word-pair rather than to use two separate
Greek words. Although several CWs point in one direction or the other,
the two factors cannot be kept apart easily.
We turn first to cases in which CWs might have occurred to the
translators of the LXX as the natural equivalents of certain Hebrew
word-pairs. In Exod 18:21 the features of the judges sought by Moses are
described by four word-pairs, Y¥2 °NIV NK WIN DPN ONT 2N -wIX, and in
Greek by dvdpac duvatouc Veo-ceBetc dvSpac Stkaloue pLacotvTag
vbTepydavlav. It would have been hard to predict which of the four
pairs of words would be represented by a CW. From the outset O°72X "x7
had a greater chance than the other three pairs because in Greek a
compound 6@eogeBr\c existed which was close in meaning to O°02X xv.4
Thus the translator of Exod 18:21 presumably used 6eo0eBhc because the
word suggested itself to him. If he had stylistic considerations in mind,
the translator would have probably also made some effort to render the
other three word-pairs with CWs. For example, ¥¥2 °x1v could have
been rendered by a combination of pco- and an additional element.
Similarly, 0°n -w3x and nox *wix could have been represented by either
existing or newly coined CWs. This example suggests that the existence
of a particular CW in Greek was often instrumental in its employment
in the LXXx.
The existence of a CW contributed to its use in the LXX when a clause
was represented as a whole by a CW, for the occurrence of such a CW
cannot be explained otherwise:
Exod 25:15 an 1907 82 Tan Vit ANT nvajD? - év Tote SaxtvAtore THe
KLBwTot EcovTat ot dvadopete d-Klyntrot
Josh 5:7 4772 gmx 199 N3>D... vw On onEK - ole "I Nootc Teprétepev
Sia TO abTouG yeyevija0at KaTa THY OSdv d-TEPLTENHTOUG
4 Cf. Gen 20:11 on>x nev - Geo-céBeia; Job 1:1 oA DX NV - Beo-aeBric.
134 CHAPTER NINE
Job 15:2819? 12¥7 87 OND NIIND] a7 Iw) - avAtabeln SE TdAELE
Eptyoug eloéOor 8 El¢ olxoug d-oLrKTOUG
Est 4:11 8727.8? IW ...7 207 DX M127 WR WWR1 wR DD - TL de GvOpwiroc
Ty yuvr oc eloedevoeTat TpdG Tov Baothéa ... d-KANTOC
The availability of a CW to the translator may sometimes be inferred
even when positive evidence is lacking. Thus é\ato-\oyéw, not attested
before the time of the LXX,° was probably known to the translator of
Deut 24:20 (qn vann). The translator of Deuteronomy probably did not
coin this CW—which does not occur elsewhere in the LXX—because the
Greek word does not exactly represent the Hebrew verb. Moreover, this
translator was not prone to use CWs for two Hebrew words. According to
the same reasoning, the majority of exegetical renderings (groups Ia,
Ia, IIIa) also must have been available to the translators.
The very existence of CWs not only induced the translators to use
them when the Hebrew and Greek words were identical in meaning;
their existence also influenced the translator to use them when the
Hebrew and Greek only roughly coincided. Thus tapa®addoo.og renders
both orn 4) (Jer 47:7, Ezek 25:16) and ov now 2y (2 Chr 8:17), and the
similar tap-GALog renders O77 )n (Deut 1:7), orn? 71n? (Gen 49:13), and
n7i1 777 (Isa 8:23). More diverse are the Hebrew equivalents of tupt-
KavoTOG: WX Ww (Isa 1:7), wx nD Iw (Isa 64:10), and wx n?DNXn TDW (Isa
9:5). This word, which was known from the time of Homer onwards,
suggested itself to the translator of Isaiah whenever he encountered a
similar Hebrew phrase.
The availability of CWs also induced the translators to use them
when they probably should not have been used. Sometimes a CW was
employed because its two constituents separately represented the two
Hebrew words well. However, when combined in a CW, the two (three)
elements formed a Greek word whose meaning differed from that of the
Hebrew.® Thus in Isa 58:9 18 937) YI¥N 72wW Av 73DIND TON ON (to point a
finger) was rendered by xetpo-tovia whose two constituents roughly
represent the Hebrew words. However, according to the preserved
evidence, xetpotovia was not used in its physical sense,’ but only
metaphorically as ‘voting by show of hands.’ Thus, although the
translator probably aimed at ‘pointing the finger,’ the readers of the
> The related éharoddyos is, however, attested in Aristophanes, Vespae, 712.
A. Debrunner, Wortbildungslehre, 58 (cf. n. 1 above) mentions similar examples of
compounds which were used in Greek literature with a meaning that was different from
the original one, e.g., mp dy{twp in Soph., Electra, 1154.
The related yetpotéviog is, however, used in its physical sense in Aristophanes,
Thesmoph., 172.
COMPOUND WORDS 135
LXX may have understood the word as ‘voting.’ Undoubtedly the
availability of the CW xetpotovia influenced the wording of Isa 58:9, a
fact which is also evident from the difference between ‘finger’ in
Hebrew and ‘hand’ in Greek. Likewise, dv-uto-vénto¢ (unexpected,
unsuspected) renders the three elements of 29 9y 92 (Sir 11:5), but the
meaning of the complete word differs from that intended by the
Hebrew equivalents of its three constituents (‘not on the mind’).
In the instances mentioned above, the existence of a certain CW
facilitated its employment as an equivalent of a Hebrew phrase. A
second factor determining the use of CWs was the translator’s stylistic
inclination to represent a closely related word-pair by one Greek
compound word rather than two separate ones. Such a stylistic motive
must have been particularly strong when the translator coined a new
compound in order to represent a Hebrew pair of words with a rendering
which, in his view, was the best one under the circumstances (see below,
section 5).
The translator’s literary taste is evident in particular in verses in
which one notes a high concentration of CWs such as
Ezek 3:5-7 0727 o-ny OX x? (6)... -7T7D7 TDW yoy ov OX RD >D
22 -wPr nxn "pin... (7) Jw? 2737) DW DY
SL4TL OV TPOG AadV Badu-yeLtr\ov Kal BapU-yAwooov ... (6) ovdé
Tpd¢ AaovG TOAAOUC AAXAO-daivouc TH dAo-yAWaadouc oVbE
oTiBapove TH yAwaoon® ... (7) ... drAd-vetxol ela.v kal oxAnpo-
Kd pool.
Three of these CWs (BaOuU-yettoc, Bapt-yAwoaoc, dd\Aé6-dwvoc) were
apparently coined by the translator of Ezekiel, a fact which underlines
the translator’s wish to employ CWs in these verses (okAnpo-Kdpd.0¢
was coined previously by the translator(s) of the Torah).
The use of ToAv-odla is also instructive in this regard. While the
Greek language has many CWs whose first element is ToAv-, the LXX
does not contain many such CWs (even though the Hebrew Bible has
many word-pairs which include a form derived from 7125). Therefore,
the literary taste of the translator of Isa 57:10 probably induced him to
coin ToAv-odta for 777 27. See further section 5 on neologisms.
8 }w? °729 is represented doubly by an exegetically motivated compound and by a literal
translation of the Hebrew. Cf. section 6 below.
136 CHAPTER NINE
2. The choice of lexical equivalents
The choice of a given CW as an equivalent for a certain Hebrew word-
pair is often found to be consistent with the lexical equivalents used
elsewhere in the LXX. For example, the two constituents of peyado-
mrépuyoc (Ezek 17:3, 7) exactly represent 07515 717) because they also
render the two Hebrew words elsewhere in the LXX when used
separately (271 - péyac; 719 - mTépvé). kako-toitjaal oe in Gen 31:29,
rendering v7 ODny nwy?, should be viewed against the background of
the stereotyped renderings of both v7 (kaké¢) and wy (trotéw). While
these equivalents are stereotyped, others reflect an interpretative
element, e.g., O° 2Nw (water-drawer) - U8po-ddpoe (carrier of water) in
Deut 29:11 (10) and Josh 9:21, and more clearly 7) 11 - pakpé-@upoc
(Prov 17:27), n&n pin - d.Ad-vetkog (Ezek 3:7), and 1121 NN - kKako-dpoovvn
(Prov 16:18).
The classification distinguishes between ‘more or less literal
rendering’ (la, Ila, IIIa) and ‘exegetical renderings’ (Ib, Ib, IIIb),
although the subjective criteria for literalness need to be refined. Some
translators were sensitive to a quantitatively identical representation
of certain Hebrew word-pairs with Greek CWs, cf. a7n 271 (a man of
great wrath) in Prov 19:19, and op mo (a woman who abandoned sense)
in 11:22, both rendered by kak6é-dpwv (of bad spirit, malignant). The
translator must have realized that he did not represent the two pairs
of Hebrew words with exact equivalents, but his literary sense led him
to preserve at least the word pattern of the Hebrew by using CWs. In
some instances this quantitative equation was apparently more
important than a faithful rendering. Thus or nu? in Job 30:25 was
rendered by d-8tvatoc, and wp1 27 in Prov 28:25 by d-trAnoTos.
3. Representation of the Hebrew
The use of CWs cannot be predicted since the translators approached
this issue in different ways. Thus combinations of the root x7 and 07778
were rendered by 8eoc0eBrc, etc. in Gen 20:11 and Job 1:1, but in the
remainder of the LXX they were rendered by combinations of goPéopat /
goBdc and 6ed¢. Likewise, 2°0°7 combined with another verb is usually
rendered in the LXX by a combination of an adverb and a verb (e.g., Jer
1:12 navn nx? - Kade Euipakac and 1 Sam 16:17 711? 2°u°n - d6pbdc
wdddovTa), but in Ezek 33:32 711 Jun - €v-dppootog and Prov 30:29 1vx
*2°0°n - Eb-SSuc¢ TopeveTtat they were rendered by CWs. 97 Awy is usually
rendered by Tovéw TO tTovnpdv, but by kako-tTroéw in Gen 31:29.
COMPOUND WORDS 137
Certain closely related Hebrew word-pairs were rendered always or
almost always with a CW. Thus 7y 7? is always represented by
OKAT|PO-TPAXTNAOG. WRI, as ‘head of ...’, ‘important’ is nearly always
dpx.- and word-pairs starting with a construct form of pay, 723, and 125
are usually rendered with a CW (see section 7).
While in these examples the close relationship between the Hebrew
words induced the translators to use a CW, some CWs rendered
combinations of Hebrew words which were not closely connected. Thus
"nh maw was rendered aptly in 2 Chr 25:12 by €-w-ypéw. Similarly, the use
of olvo-rotéw for 7° mnw in Prov 31:4 and of USpo-totéw for nw o°n in
Dan 1:12 takes advantage of the existence of these compounds in Greek,
while these combinations were rendered elsewhere by tlvw USwp and
tlvw olvov. The same applies to CWs which represent a whole clause.
In all these cases the use of a CW was determined by its existence in the
Greek language rather than the essence of the Hebrew word-pair.
Greek CWS represent various types of combinations of Hebrew words.
The most frequent type in Hebrew is a combination of a noun or adjective
in the construct state together with a (second) noun. The following types
may be recognized:
(a) construct adjective + noun = adjective
e.g., WY 1)? - oKANPO-TPaXTAOG Exod 33:3
(b) construct noun + noun = noun
e.g., 225 nq - okAnpo-kapdla Deut 10:16
(c) noun + adjective = noun
e.g., 190 NN - OALyo-ywuxla Ps 55 (54):9
(d) noun + adjective = verb
€.g., TWNDI NN - okv8p-wralw Prov 15:13
(e) noun + negation = adjective
e.g., 22 PR - d-KdpStoc Jer 5:21
(f) verb + noun = verb
€.g., V1... WY - KAKO-TOLEW Gen 31:29
(g) verb + verb = verb
e.g., NWY? NDA - peyado-oLréw Sir 50:22
(h) verb + preposition = verb
€.8., “19 10 - Exk-pevyw Job 15:30
(i) verb + negation = verb
e.g., VOW NX? - d-TELbEw Josh 5:6
(j) clause = adjective
e.g., 0? 12M" X27 O'NI - olKouG d-oLK| Tove Job 15:28
(k) preposition + noun = noun
e.g., Onn 2y - ém-odpayrCdpevor Neh 9:38 (10:1)
138 CHAPTER NINE
In all of the examples, the translators recognized the two elements of
the Greek CW, each of which represented a separate Hebrew word.
Such a linguistic analysis of the Greek word is natural in most cases, but
in some it indicates a refined linguistic understanding. Thus the
translator of Prov 21:15 showed knowledge of the etymological
background of kak-o-tpyoc by using this word as an equivalent of px ?y».
The same refers to d-a8evéw (to lack power, to be weak) for n> 7°X in Isa
44:12 and d-xovotwe (against the will) oyp x72 in Sir 25:18 (3).
4. The approach of the translators to the use of CWs
Some internal evidence indicative of the translators’ approach to the
use of CWs may be inferred from the distribution of CWs in the books of
the LXX as well as from a few selected examples.
From a formal point of view, the representation of two Hebrew
words by a Greek CW may not be considered literal, even though from a
linguistic point of view the CW may be the best possible rendering of
the Hebrew. According to this principle, more CWs may be expected in
the freely rendered books than in the books which were translated
literally. Indeed, CWs which represent more than one Hebrew word
occur more frequently in the freely rendered books of Job, Proverbs, and
Sirach than in other books (see section 7). However, the evidence may
be misleading since the Hebrew text of these books, particularly that of
Proverbs, may provide more opportunity for the use of CWs than other
books (the sapiential literature seems to contain a relatively larger
number of closely related word-pairs than the other books).? On the
other end of the scale, the literal translation of Samuel-Kings (large
sections of these books belong to kaige-Th) contains few examples of
CWs representing two Hebrew words. Undoubtedly the Hebrew of these
two books provides more opportunity for the use of CWs than the few
that they actually contain.
The evidence from Aquila and Symmachus points in the same
direction as the evidence from the LXX. The literalist Aquila rarely
used CWs for more than one Hebrew word except for some CWs starting
with an alpha privative;’° on the other hand, Symmachus, who sought
9 That is, construction of a participle with a noun, and also combinations of a noun or
adjective in the construct state, together with another noun. To the best of my knowledge,
the relative frequency of such constructions in the various books of the Bible has not been
investigated.
10 Fg. Jer 17:11 vpwn2 xd(0) - dkpitel. See J. Reider and N. Turner, An Index to Aquila
(VTSup 12; Leiden 1966).
COMPOUND WORDS 139
to produce an elegant translation, strongly favored CWs of the kind
described here, as was noticed by Field."
The description of the two different approaches to CWs is supported
by some parallel renderings of Hebrew word-pairs in rather free and
rather literal translation units:
Dan 1:4 4x70 7110 - eb-eLdete Th: kadoug TH Swe
Dan 1:12 Anwi o°n1 - kal DSpo-toTtetv Th: kal b&wp TLépE8a
Dan 3:29 (96) 12v anx* (Q)-BrAac- — Th: eit BrAaodnyulav
Th
Th
PnUTON
Dan 7:19 n° 19°74 - Ure p-ddBou
Dan 11:27 11277 31D - Pevdo-
MOYTGOVOL
Judg 10:16A.. 1w53 AXpM - Kal B..: Kal MALYWEN F wuyx7] avTod
WALYO-WUYNOEV
: POBE POV TIE PLOGWG
: WEVST AAATOOVGL
Judg 16:3A.. 92°97 -3n23 - Tept B..: €v floer THIS vuKTdOG
TO WEDO-VUKTLOV
Judg 16:26A.. 1172 p-1nan - TOV B..: Tov KpaTobvTa Tv xetpa
xelp-aywyotvrTa avtév auTov
2 Kgs 22:14 ona nw - Tot 2 Chr 34:22: puAdaoovoay Tac
LuaTLo-@vAaKoOG atoAd¢ Rahlfs; MSS: évtoAdc)
5. Neologisms
Some of the CWs listed in section 7 are designated as ‘neologisms of the
LXX.’ A neologism of the LXX is a Greek word which, to the best of our
knowledge, was coined either by the translators of the LXX or by a
previous generation, in order to express biblical words which, in their
view, could not be expressed adequately by the existing Greek
vocabulary.!* Neologisms are either compounds which use elements
existing in the Greek language or are derivatives of known roots. New
roots were used only when Hebrew words with Greek endings were
introduced into the vocabulary of the LXX, e.g. cafBBatl{w, ‘to keep the
Sabbath,’ derived from od8S8ata = naw.
The term neologism must be used with caution since most neologisms
are composed of elements which were previously known to the
translators and their contemporaries. Therefore, the German term
11 Field, Hexapl., xxxi. E.g., 1 Sam 25:3 95u naw - eb-Siavéntos (LXX: &yah) ouvéoe). Isa
52:8 pyr yy - ddBadto -pavuic (LXX: dp8aryol mpd¢ dp@adpovc).
The most extensive discussion of neologisms is found in Lee, Lexical Study. See also K.
Hartung, Septuaginta-Studien, Ein Beitrag zur Griicitat dieser Bibeliibersetzung (Bamberg 1886)
22-45.
140 CHAPTER NINE
Neubildungen is more precise. For example, although Baétc and dwvy
were well-known words, Ba0v-dwvoc was apparently created by the
translator of Isa 33:19 to represent 1Dw *pny. There is another reason for
a cautious use of the label ‘neologism’: a word described as a neologism
on the basis of our present knowledge may, in fact, be contained in an as
yet unpublished papyrus fragment or the word may never have been
used in written language. Although the assumption of a neologism must
remain tentative, the probability increases when the assumed neolo-
gism is a precise replica of the Hebrew (e.g., weyadd-capKog = 72 773),
or when it is characteristic of the Hebrew language (e.g., oxAnpo-
Tpdxndoc = Wy 7?) or of the Hebrew Bible (e.g., 8 utepo-vdptov = mI1wN
71n).
In the list in section 7, CWs which are not attested before the time of
the LXX are denoted with a plus sign (+). The basis for this designation
is the vocabulary recorded by LSJ and LSJ, Supplement. Not included in
this group are words which merely received a new formative element in
the LXX; e.g., olvo-totéw (= 77 nw Prov 31:4) is attested first in the
LXX, but the identical olvottotd¢€w was known from Homer onwards.
The relatively large number of neologisms among the CWs may be
due to the fact that CWs were formed during all periods of the Greek
language, including that of modern Greek. Similar to other Indo-
European languages, it is characteristic of Greek to enlarge its
vocabulary by coining new CWs to express situations and objects which
were previously unnamed.
Nearly all the CWs which are designated as neologisms contain at
least one element on the basis of which other Greek words are formed as
well. Thus the use of pakpdé-Oujoc¢ for O°DX 778 must be seen in the light
of the fact that pakpdc¢ is used as the first element of CWs from the
early days of the Greek language. Similarly, the examples of group II
in section 7 contain at least one formative element (viz., a preposition),
and most of the words in group I contain a formative adjective or adverb
(e.g., €U-, Bapu-, pakpo-, eyado-). The same principle applies to group II
(e.g., KaKo-Tolew). Among the productive elements in this group one
finds both verbal elements (mainly in final position) such as -yevr\c and
-ToLéw and nominal elements (mainly in initial position) such as 7\TaTOo,
dveyo- and tveupato-. Very frequently both the first and second
constituents were productive, e.g., peyaho-Troréw and dveyd-d8opoc.
Groups Ia, Ia and IIa (‘more or less literal renderings’) contain more
neologisms than groups Ib, IIb and IIb (‘exegetical renderings’). As
expected, the translators coined new CWs in order to represent
COMPOUND WORDS 141
combinations of Hebrew words which were equivalent with the Greek
more easily than CWs which reflected exegetical renderings.
The recognition that a certain CW is a neologism underscores the
translator’s determination to represent a given word-pair with one CW
rather than two separate words. This inclination manifests itself in
particular in exegetical renderings. Thus the translation of 22 no1y by
okAnpo-KkapS(a in Deut 10:16 conforms with the translation of the
parallel stich, T1y wpNn RX? ODD 191 - kal TOV TPdxNAOV byGV ov OKANPL-
vette. Similarly, 5wpo-ArjmT¢ (a receiver of bribes, cf. Supa Aap Bdvw
passim in the LXX) as a translation of 982 9812 in Prov 15:27 conforms
with the translation of the parallel stich, 77m ninn xiiwi - 6 6 ploy
SwWpuwv AHpWere od cera.
This determination manifests itself also in CWs which are unusual
within the literary framework of the LXX even though they conform
with the pattern of forming compounds in the Greek language. This
pertains to such CWs as wy 3?2X - onTé6-Bpwtoc (eaten by moths) in Job
13:28 and 1950 7pw - xapako-Bodla (forming a palisade) in Ezek 17:17.
6. Double representation
Sometimes one of the words of a Hebrew word-pair is represented both
by one of the constituents of a CW and by a separate Greek word. In
some cases the translator may not have realized that these words were
represented twice, while in other cases the double representation may
have resulted from a hypercorrection inserted by a reviser.
(a) IND m2 was rendered in Gen 39:22 and 40:3 by S€apwTptov, and its
master, 10 N32 Ww, was rendered in 39:21, 22 by dpxt-SeapodtAaE based
on the similar S€opwdvddKetov.}! In v. 23, however, he is named 6 dpxt-
SeopopvAaE tot S€opwtnplov. In this case, 770 n2 is represented twice.
(b) In Deut 17:18 771n 7iwn was rendered by Seutepo-voptov. In Josh
8:32, however, 771n in the same expression was rendered doubly: 73uvn
mw nan - TO S€utepo-véucov, vdpov Mwuor (om. vépov LXX° Copt). This
rendering may have resulted from the translator’s avoidance of a
rendering like TO Se€uTepovéytov Mwuon.
(c) The stereotyped rendering of X28 Ww is dpx.t-oTtpdtnyosc. In a few
cases, however, X28 was doubly represented: Gen 21:22, 32 ®:kod 6 dpxt-
OTPaTHYOG Tic SuvdpEws avtot; see further Josh 5:14 and 1 Chr 19:16, 18.
13 Cf. further SecpoptAae (gaoler); both words are absent from the LXX.
142 CHAPTER NINE
(d) In Gen 17:12 (na 179° - olko-yevtic Tie olklag gov), n-2 was repres-
ented twice (the last three word in the LXX are omitted by manuscripts
MO‘!7' 135 C’"", ete.)
(e) In Isa 35:4 (29 171 - dALyd-puxot TH S.avolq), 19 was represented
twice.!4
7. The classified evidence
The purpose of the lists is (1) to provide an inventory of CWs
representing two or more words in the LXX,'° (2) to record the only or
the major Hebrew equivalents of the constituents of the CWs, and (3) to
remark on peculiarities of the occurrences of the CWs in the LXX
(hapax legomena, etc.) or in Greek literature (neologisms).
The lists contain CWs which represent two or more Hebrew or
Aramaic words found in books whose Vorlagen are known. The
examples are culled from the concordances of Trommius, HR, and
Smend!¢ and are recorded on the basis of the critical editions of the
LXX.
Numerals and word-pairs containing a numeral (e.g., O°wIN AVIw -
ETTad-yNvov (Ezek 39:12); D-avD AND - EkatovtTa-TAdotov (2 Sam 24:3) are
excluded from the lists because they must be dealt with separately.
Other words which are not included are compound Greek prepositions,
such as tap-€€ representing 19 72° in Judg 8:26B and Sir 49:4 and tnep-
dvw representing ?¥ }7"?y in Deut 26:19. These have been omitted because
of the difficulty in assessing the value of the elements of such CWs.
Since the concordances provide incomplete data with regard to the
CWs, the lists are not complete, although in reality they are probably
nearly so. Furthermore, it is not always clear from the manuscript
evidence whether two words form a CW or are separate words. !”
14 See further n. 8 and the following renderings: Gen 46:32, 34 7170 -w1in - dv&pec KTHVO-
Tpddo.; Hos 9:7 N17 wx - GvOpwtog 6 mvevpato-pdpoc; Zech 3:8 now *wix - dvSpec tepato-
okétrot. In these three examples, w*x is probably not represented doubly, but the CW
reflects a free rendering. In the first example, in Gen 4:20 and Num 32:4, 71? alone has
been rendered by ktmvotpdégoc . |
Greek words which cannot be regarded as CWs are not included even though their
morphemes represent two separate Hebrew words, such as ttepwtdéc¢ - 919 Dvr (Prov 1:17)
and 6upddne - 4x 2ya. The same refers to similar Hebrew words such as o°n)wn whose
constituents are represented by 6.-youla in Judg 5:16, possibly under the influence of the
rendering of xwn by yép0¢ elsewhere in the LXX (Exod 23:5; 2 Kgs 5:17).
A. Trommius, Concordantiae graecae versionis vulgo dictae Septuaginta interpretum
(Amsterdam /Utrecht 1718); HR; R. Smend, Griechisch-syrisch-hebraischer Index zur Weisheit
des Jesus Sirach (Berlin 1907).
17 See Rahlfs’ remarks on Num 5:20, 29 wx nnn- wt’ dvipde (‘Uttav&po¢ editiones’) and
Ps 48(47):3 no - eb picaiv / eb-piCav; cf. Psalmi cum Odis (GOttingen 1931) 157.
COMPOUND WORDS 143
All Hebrew equivalents and scriptural references are provided in group
III and almost all in group II. Some references are omitted in group I
because the aim of that section is to list the Greek evidence together
with the Hebrew equivalents of the first constituents of the CWs.
The following notations are used:
* - The CW occurs only in the mentioned verse(s) in the LXX. If not
indicated otherwise, such an asterix denotes a hapax legomenon.
+ - The CW is not attested before the time of the LXX (see section 5).
only 2 Heb. w. - The CW represents only combinations of two words, e.g.,
d-KdpS.tog = PX 27,29 PR, and 27 70n.
et al(iis locis) - The CW occurs in the mentioned verse and also
elsewhere.
only Sir (and sim.) - The CW occurs only in Sirach.
no remark on the frequency of a CW - The CW occurs in the mentioned
verse as an equivalent of two or more Hebrew words, and also elsewhere
as an equivalent of single Hebrew words.
CWs are classified according to internal Greek criteria!® or according
to the Hebrew word-pairs which they represent, while within each
group information is provided relating to the techniques employed. The
classification has been simplified with regard to possible subdivisions
and to the linguistic terminology used for CWs.
The CWs are divided into three groups; in each group the
constituents of the individual CWs relate to each other differently:
(I) CWs containing an attributive element;
(II) CWs whose constituents are subordinated to each other;
(III) CWs whose first constituent is a preposition (preverb).
Each of the three categories is subdivided as follows:
(a) more or less literal renderings;
(b) exegetical renderings.
I. CWs containing an attributive element
In the CWs included in this group, the first constituent describes the
second one. The two Hebrew words are generally related to each other
in the same way as the Greek elements, e.g., ]1w? 725 - Bapt-yAwooos
(Ezek 3:5). The first constituent of the CW is either an adjective (e.g.,
1}50-dwvoc) or an adverb (e.g., €b-mpdowtoc). The two constituents of the
18 CE A. Debrunner, Wortbildungslehre, 15-83 (cf. n. 1 above); E. Mayser, Grammatik der
griechischen Papyri aus der Ptolemderzeit I, Wl (2d ed.; Berlin/Leipzig 1936) 153 ff; E.
Schwyzer, Grammatik I, 425 ff.
144
CHAPTER NINE
CW appear in the same sequence as the Hebrew words, e.g. 75 1725 -
Laoxvd-dwvoc.
a. More or less literal renderings
d(v)-
Ba®u-
Bapvu-
Bpadv-
d€ UTEPO-
PR 2D ~2a(a) on pon 8? na? any
e.g., d-UVNOTEUTOG* TWIN XN? WR (Exod 22:15)
further: d-BaoiX\evtog*, d-BorOntoc*, d-6idAvTOG", a-
Kalpwe*, d-Kdp8.oc (only 2 Heb. w.), d-katdtotoc*t,
d-KavoTo<g", d-Kivntoc, d-KAnTocT, d-Kovalwe, dv-
aplO@untoc, dv-eێXe yk Tog (only 2 Heb. w.), dv-
e€txviacotoct (only Job), dv-hkooc, dv-latoc, dv-uT0-
vor|T0G*, av- whedAjs (only 2 Heb. w.), d-olkntoc, a-
talSevutos, a-1eLbéw, d-tepltuntoct, d-onttos, d-
obevéw, d-Tel xLaoToc, a-fEeLbuic*, d-dpoovvn, d-dpwv,
d-ypNoTOG
POV
Babv-dwvoc*t TOW pny Isa 33:19
further: Ba8v-yetXoc = (ADw pny)
i pe)
BpadSv-yAwooog*t Ww? 72d
further: Bapv-KkdpStoc* (27 32D Ps 4:3; MT diff.)
TAD
Bpadt-ykwoooc*t pw> 72D Exod 4:10
WW
SeuTepo-véputov*t min imwn Deut 17:18; Josh 8:32
HO a?
€U-TPdGWTOG IN 11D" Gen 12:11
further: €v-dpyootoc*, ev-eLbrj\c*, ev-dbwc"*, ev-prCaiv*,
Ev-hpoatvn
rai
480-dwvog* Diy iD? Ezek 33:32
ald
Laxvé-hwvog 7m) 322 Exod 4:10 (only Exodus)
gages
hew-Te Tplax+ voomnx Ezek 24:7, 8; 26:4, 14
FAR AI
Wakpd-BLoc* obs age puma Isa 53:10
further: waxpo-niepetw (only 2 Heb. w.), paKpo-
fpepoc** ylvouat, paxpo-Oupéw*t, pakpo-Supla
Le yado-
VEO-
vw8po-
dXLYO-
dd0-
dCu-
TOAU-
Tpau-
oK\npo-
OTEPEO-
TATIELVO-
vy Ao-
COMPOUND WORDS 145
(only 2 Heb. w.), paxpd-6upL0G, pakpo-xypovl (wt, paKpo-
xpdvriog ylvopat (elt) (only 2 Heb. w.)
27
we yadd-capKkoc*t Awa 371 Ezek 16:26
further: weyado-1Tépuyoc**
TNX WRT
veo-unvla wan? TNX Exod 40:2, 17
VIN WN Num 10:10; 28:11
V2
vwOpo-KdpS.oc*t 95 711 Prov 12:8
TP ,2Dw
dALyo-Wuxe Ww wD) AXP Num 21:4
further: d\Ly6-BLoc*, dAvyo-yuxla (only 2 Heb. w.),
dAvyo3puxoct (only 2 Heb. w.)
25
dd\0-KapTdopar*t “WOiaa- PPD Sir 45:14
ak ey
d¢v-BvpL0G* MDX ANP Prov 14:17
To 7so
TOAU-€ Ae og + ton 273 (only 2 Heb. w.) Ps 144:8
further: ToAv-rLE pevw*, ToAU-TpLEpos ylvouat (only 2
Heb. w.), todv-oSla** todv-ppywv*t
Ow
Tpav-Ouoc t nin>pw Prov 16:19 (only Prov)
JIT ,WPY ,TW/P(D)
OKANPO-TPA XTNAOG WY Twp Exod 33:3 et al.
further: oxAnpo-KkapSla**t, oxdnpo-Kdp&tocg (only 2 Heb.
Ww.)
Pin
oTepeo-KdpS.o0c*F 35 pin Ezek 2:4
2OW
Tate.vé-ppev-*t min ow Prov 29:23
dh
vynro-KdpSiocg ** a? fot Prov 16:5
b. Exegetical renderings
a-8{Kwe
a-80vaTog
d-Kovolwe
dd\A6-yAWOCOG
ddA6-hvAOC
win D2 Job 24:10
oY mwp Job 30:25
ovDy X22 Sir 25:18(3)
yw? 7D (?) Ezek 3:6
15) 72 Isa 61:5
146
&AS-duVvoG*t
a-hoyog
d-LéETPHTOG
dudote po-8€ Etoc*t
dv-ardyc
dv-é\TLOTOG*
a-tral Se€vtoc
a-TAnotla*
d-TANOTOG
di-wWpoc
8{-yAwoooc
EOXATO-yT] pwc
ev-adoc*
€v-086W
€U-Tpoot) yopoc*
EUPU-XWPOG
Opacu-Kdpdtoc*
Loxvd-duvoc*
KaKo-ppoatvyn*t
KaKO-dpwv*
waKpdo-OuLLoG
we yadd-Bowv*
ILLKpO-AO'Yor *
udv-opxic*t
dALyO-WuyKe w
dALYO-uXla
ddryd-buydgt
61L0-{17}T pLoc*
VLLO-TAT plo *
ok\npo-Kapéla*t
oKv0p-wid Cw
xpnoto-ybera*t
CHAPTER NINE
TOW poy (7) Ezek 3:6
o-now 327 Exod 6:12
a7? 9An4 Isa 22:18
re ia alla |2) Judg 3:15; 20:16
wi TV Sir 40:30
iP W? Isa 18:2
JIN WR Sir 8:4
7010 89 Sir 37:31
wor 5y3 Prov 23:2
wd) 2m Prov 28:25
con aw Isa 65:20
ony 3ya Sir 514-6:
WID WX Sir 41:2 (only Sirach)
O-NSW wR Job 11:2
jn -now Sir 6:6
77743: an1(7) Gen 24:27, 48
ow Raw Sir 6:5
Be? amr Judg 18:10A
27 110 Prov 14:14
a 70n9 Prov 21:4
o-now 27 Exod 6:30 (cf. 4:10 LXX)
mn 72) Prov 16:18
Tan 37} Prov 19:19
ovo no Prov 11:22
At lp Prov 17:27 (see also Ia)
aryan Prov 21:4
op 29 Sir 14:3
Jer mn Lev 21:20
ni pip Sir 4:9
WY? 7 Hab 2:13 (cf. Judg 8:4A)
MYO 17 Ps 55(54):9
A Ay Isa 54:6
VAX 73 Gen 43:29
ph wamirhla Lev 18:11
99 n>37y Deut 10:16
MNDI MN Prov 15:13
2) | Sif 37511
II]. CWs whose constituents are subordinated to each other
The two constituents of the CWs included in this group relate to each
other in various ways which may be described as sentences in which one
COMPOUND WORDS 147
element relates to the other in either the genitive, dative or accusative
case. Thus a S0pato-ddpoc is someone who carries (dé pet) a lance (Sd6pu,
Sépatoc), and ontdé-Bpwtoe is something consumed (-Bpwtoc) by moths
(ofc, ontdc).
In most of the items, the sequence of the words in the Hebrew word-
pair is reversed in Greek in accordance with the rules of composition of
Greek CWs. Thus the ‘cutters of wood’ (o7x¥ 720n) of Deut 29:10 are
‘wood-cutters’ in Greek (EvAo-Kétrot), and the ‘drinking of wine’ of Prov
31:4 is ‘wine-drinking’ (olvo-totéw) in the LXX. The components occur
often in Greek in both sequences,’? including in the LXX where one finds
both dpxl(-dudoc (vIv WN; Deut 29:9) and }vA-apyoe (baw 771; Deut
31:28). yAwooo-xaptTéu (71w> p"2ND) is not attested in Greek literature
outside Prov 28:23, but xaptTo-yAuooéw is known in classical Greek.
Some of the adjectives listed as constituents of CWs in group I, are
also listed here, though in a different syntactic function. For instance,
the aforementioned peyado-1Tépuyoe (with great wings) should be
compared with peyado-trovéw (to do great things) listed below.
a. More or less literal renderings
dveyd-dbopoct op yyw Gen 41:6
dpyup-uvntog NOD npn Gen 17:12 et al.
apXL- WRI Ww
e.g., dpxt-SeapopvAaé*t - ano manw = Gen 39:21
further: dpyt-Seopastnc**, dpx.-euvotxoc* (only Daniel) dpxt-
udyetpoc (only 2 Heb. w.) dpxt-otvoydoc*t, dpxt-tratpiitye**,
dpxt-o.totroréc**, apxt-oTpadtnyoc (only 2 Heb. w.), dpxt-
owuatodvAak*t, dpxl-dudoc*
BuBAto-6HKj x°790 m3 (only 2 Heb. w.) Ezra 6:1
yAwooo-xapiTéow*t yw? p-?nn Prov 28:23
8upeo-ddpoc* aX Nw] Wehr i225
Sopato-ddpoc* nn... XW} 1 Chr 12:24
édato-hoyéw*t ony Dan Deut 24:20
évSo-yevyc*t mantzin Lev 18:9
C-w-Ypeéw naw 20chr 252
{trato-oKoTré opar*t 3232 7X7 Ezek 21:26 (21)
Geo-cé Beta ODN ANT Gen 20:11 et al.
§eo-cE Bc onbox xv Job 1:1 et al.
Bupeo-ddpoc*t TY NWI LChEI2 25
19 cf. Debrunner, Wortbildungslehre, 80 (cf. n. 1 above); e.g., d€p-aamig / domtSy-pdpor,
dpxl-trodr¢ / wodl-apxoc.
148
Luatio-dtAak*t
lt1m-d pxat
KQKO-TTOLE W
KQK-0-UPYOG
\LO-o-upyiKkda *t
wayado-troLéwt
LE-Yao-ppn Love w
Leyado-pprjev*t
[LEOO-VUKTLOV
Ev\o-KéTr0G*
Evdo-dopla*t
680-TOLEW
olko-yevyc
olko-vé,L0G
olvo-TroTé w
dLvO-T6TNG*
drA0-pdpoc*
TAT PL-d pPxXNG
TETP6-BOAOL
TVvEvLAaTO-dopéopat**
TpWwTo-Babpé w*t
TITE PO-ue w*
Tupl-KavoTog = wx wv and sim. (only 2 Heb. w.) Isa 1:7
onté-Bpwtoc*t
vép-aywyol
vVSpo-TIOTE w*
v8wo-ddpoc*
gdir\-apaptyyov*t
further: dtA0-yewpydc*, diro-yvvatocg*
dvA-apxoc*
xapako-Borla*t
XELp-aywyéw
xelpo-tovla*
xAwpo-Botdun*t
Xpuao-TépeuTog**
xpvoo-xyd0c
WadL-wddce +
wevbo-hoyé w*
CHAPTER NINE
o°72 Tw 2 Kgs 22:14
o-w7 °2y2 2 Sam 1:6
7 nwy Gen 31:29
pS 2VD Prov 21:15
Jax nwan Exod 31:5
Mwy? NPA Sir-50:22
19 2-17 Ob 12
aie ap aa bee Ps 12(11):4
na? an Judg 16:3
oxy Jon Deut 29:10; Josh 9:23
Oxy Jaq Neh 10:35
707 7p Isa 62:10
Mat Jer 2:14 et al.
eee Be Qoh 2:7 et al.
Man OY WR 2 Kgs 18:18 et al.
jae eae Esth 1:8
7 anw Prov 31:4
7? 820 Prov 23:20
MX NWI 2 Chr 14:7(8)
MAX WN 1 Chr 24:31 (only Chronicles)
Vp STON Job 41:20
mT yNw Jer 2:24
2YD NOD OW Esth 3:1
Tax Dy Isa 40:31
wy W2K Job 13:28
O° "NS Isa 41:18
rinw arn Dan 1:12
O° INW Deut 29:11(10); Josh 9:21
Ywd INN Prov 17:19
Daw 771 Deut 31:28
1220 DW Ezek 17:17
A ace iin Judg 16:26A
YIRN Nw Isa 58:9
NWT py 2 Kgs 19:26A
mW Dit Exod 25:17(18) B*ob>
2772 Isa 40:19
re myi Sir 47:9 (only Sirach)
AD ttle Dan 11:27
COMPOUND WORDS 149
b. Exegetical renderings
Brao-pnuéw YW TDN Dan 3:29(96)
BrAdo-pTL0¢ * PX yA Isa 66:3
YN-YEVHS OTN 73 Ps 48(49):3
Swpo-AyamMs*t YI yE Prov 15:27
KaKO-TOLE W m2 yy Prov 6:18
KAOTIO-Popé w* aip ane) Gen 31:26
ALVO-KaAd Ly yy “nwd Josh 2:6
LLeo-rUBpla own onD Gen 18:1 et al.
UNTPd-TOALG 1370 Ty Josh 10:2
O6TrAO-Ud XOG* ovt *2D Isa 13:5 (cf. v. 4)
dp0aryo0-daviuig* pny 25227 Esth 8:13
TOVTO-Popé w* apeeiels Prov 30:19
TIPLOTOPO-ELEHG*T nyvap >a Isa 41:15
dbavTaoLo-KOTIé w”* NTN) W Sir 4:30
bird -vEeLKoc* AX PMN Ezek 3:7
xopto-pavéw*t am iaaloe Prov 24:31
Ill. CWs whose first constituent is a preposition (preverb)
Many Hebrew words are best expressed in Greek by prepositions or
preverbs. E.g., the idea expressed by XW] in O°17y NW) is suitably represen-
ted by dva- as in Deut 4:19 kal py dva-Brédac (ele Tov ovpavév). 75n3
NoD2 is aptly rendered by tTept-npyupwpévat in Ps 68(67):14. A certain
pattern in the representation of Hebrew verbs and prepositions by
equivalent preverbs is recognizable:?
dva- NWI wn
atro- ND NP Dw
éK- 1D
ém(t)- "7NN DY
Tapa- ~1D?
TE PL- 770
Tpo- 1D?
Ouv- oy
UTIO- nnn
UTE p- DY
CWs whose first constituent is a preposition (preverb) are listed below.
Words of the type elotropetopat et¢ - 2X 727 are not included because on
a formal level such CWs represent only one Hebrew word. The
20 cf, Margolis, “The Greek Preverb” (see n. 1 above).
150
relationship between these CWs and the CWs under review must be
CHAPTER NINE
investigated in a separate study.
a. More or less literal renderings
dud.-Bodetic*
dva-Bodw
dva-Bvé Tw
dv-ayyédAw
dva-KUTTW*
dva-tvé w*
AVTL-TPGOWTOL
ad T-ava_oyuvTéw*
aTO-YLVWOKW*
dtro-Kpl vod
do-\LO6w*
Sta-Tapbevevw*
Sta-p0el pw
éEK-pEevyw
év-O0pniua
€v-oTTA0G*
&év-opkoc*
EV-OXOG
é1-aKoA\oudé w
étr-odpayl(w*
KaTa-xypvoea*
Tapa-BarAd OOLoG
Tap-aKovw
Tapd-KeLwaL*
Trapa-TlOnpie
Tap-d ALOG
TE pL-apyupdw*
TeEpL-ovuxlCw*t
nanan wry
Dy? NWI
Oy NVI
O'1D NVI
927 wr
WNT RWI
ny awa
o71D 2y
02371 IND
YT RN?
27 wA
72ND On
Aen Sa wy
Tex 357
"1 NO
32°90 120
o-732 wl2%n
miaw 22
WR12 O7
“ANN NIN
pinnn oy
ant *4
on yn
oa now 9y
WY TR
aT ICLA
~1D? INI
BLS ra a
*199 Jn
"D0 ow
on yn
on yn?
ay a la.
MOD] WN}
02 pwn
DOD 71D
ONDE AWY
Isa 19:8
Gen 21:16
Deut 4:19
Job 22:26
Gen 37:14
Job 10:15
Job 9:18
Ezek 42:8
Jer 3:3
Deut 33:9
Sir 11:8
Exod 15:16
Ezek 23:3, 8
Sir 47:22
Job 15:30
Sir 35(32):12
1 Kgs 22:10
Neh 6:18
Josh 2:19
Josh 14:14
Neh 10:1 (9:38)
Deut 1:1
Jer 47:7; Ezek 25:16
2 Chr 8:17
Esth 3:8
Sir 30:18 (see also next entry)
Sir 31(34):16
Sir 15:6
Gen 18:8
Exod 19:7
Deut 1:7
Gen 49:13
Isa 8:23
Ps 68(67):14
Exod 38:17
Isa 30:22
Deut 21:12
Te pi-TLOnLLL
TE PL-Ypvodw
TpO-avaTé AAwW*
Tpd6-TATTTOG*
TPO-TTOpE VOLLAL
Tp00-aTroOvAaKw*t
TpPO-TpE Yw*
ovy-ylvoua.
ovuy-KOLTOG*
oup-Bovdla
ovp-BovrA0G
oup-trlyw*
OuLL-TOve w*
ouv-S€iTIvé w
obv-S5eLTIVvoG"*
OUV-ELOE PXOLAL
OUV-OLKEW
OvUV-Tpé xw*
OvV-wLdTN]G*
av-oKnVvoc*
Utr-avdpoc*
UTE P-KELLLAL
UTE p-hoBog *
b. Exegetical renderings
dvtT-akovw*
eELa-o850¢
E€v-5ELA.
KaTa-BLdw*
KAT-OLKLCOpaL
Ta pd -VOLLOG
TE PL-TOLE OLLAL
TE PL-OTOLLLOV
TE pl-xWPOG
COMPOUND WORDS
370 ow
qt nDdon NTDX
nox qu?!
IN 7ON
"19? 720
nv Oh
"ID? PY
py am
an Nav
TRY WR
TRY WR
710 DVI
ov anw
oy on124
on? nx an?
on) dua
ov NI
ON NI
ay pin
Peg ava
jaa be aa
DANI WR
wer nnn
Dy 29
my anaes a
mY xX?
o-nnd x1
a7 on
on? 92x
naw> Sx
D793
WN WR
nwy nam
WRIT °D
oy niz0n
M377 Aw
Exod 40:8
Isa 30:22
Ezek 17:9
Exod 10:6
1 Sam 17:7
Exod 21:29
1 Sam 8:11
Gen 39:10
Mi 7:5
Ps 119(118):24
Isa 40:13
Sir 6:6
Esth 7:1
Sit 3/25
Prov 23:6
Sir 9:16
Esth 2:13
Deut 22:13
Ps 50(49):18
Gen 14:13
Exod 3:22
Exod 16:16
Num 5:20, 29
Prov 31:29
Dan 7:19
Job 11:2
Prov 8:3
Prov 10:21
Amos 7:12
Exod 2:21
Judg 19:22
Prov 6:12
Jer 48(31):36
Exod 28:32
2 Chr 16:4
Esth 9:12
21 Cf. the frequent rendering of nox by dvaté\w. "D1 of MT is probably reflected as nov.
This rendering reflects an unusual understanding of the Hebrew phrase of Sirach.
2
TPO-QULWG
oup-Brdoua
CULL-TOOLOV
UTré p-8upov*
UT6-yaLog*
CHAPTER NINE
337 2 Chr 29:34
ov r727 SIF 13:5
TP aamnwn Esth 7:7
O°D)071 NIN Isa 6:4
ek nnn Jer 38(45):11
CHAPTER TEN
MIDRASH-TYPE EXEGESIS IN THE SEPTUAGINT OF
JOSHUA
The critical investigation of midrashic elements in the Greek trans-
lations of the Bible began with Z. Frankel and reached a state of
refinement in additional studies, especially by L. Prijs and D.W.
Gooding.! No special attention has been paid in this regard to the LXX
of Joshua, which is analyzed here.?
A few words of clarification are in order on the nature of the
midrashic elements to be discussed. Those elements are considered
midrashic which deviate from the plain sense of MT and either reflect
exegesis actually attested in rabbinic sources or resemble such exegesis
but are not found in any midrashic source. Further study is needed in
order to solidify our criteria for the identification of Midrash-type
exegesis in the LXX.
A special problem arises with the question of the relationship of the
midrashic elements to their Hebrew Vorlage: when a certain element in
the LXX is recognized as midrashic, only its content and not its
background is characterized, for the midrashic exegesis may have
entered the translation on either the Hebrew or the Greek level (the
1 Frankel, Einfluss; J. First, Spiiren der palistinisch-jiidischen Schriftdeutung und Sagen in der
Uberstezung der LXX, Semitic Studies in Memory of Rev. Dr. A. Kohut (Berlin 1897) 152-166; L.
Ginzberg, “Die Haggada bei den Kirchenvatern und in der apokryphischen Literatur,”
MGW] 42 (1898) 537-550; 43 (1899) 17ff.; V. Aptowitzer, “Rabbinische Parallelen und
Aufschliisse zu Septuaginta und Vulgata,” ZAW 29 (1909) 241-252; Prijs, Tradition;
Gooding, “Text and Midrash” and the earlier articles of Gooding quoted there: id., Relics.
Additional literature until 1948 on rabbinic exegesis is mentioned by Prijs, Tradition, xiii and
105. See also the relevant discussions in the monographs on the LXX of the Torah, Isaiah,
Daniel, Job and Proverbs.
In addition to the literature mentioned below, see D.W. Gooding, “Traditions of
Interpretation of the Circumcision at Gilgal,” Proceedings of the Sixth World Congress of Jewish
Studies (Jerusalem 1977) 149-164 in which the background of the LXX of Josh 5:4—5 is
described as rabbinic (cf. Shir Hashirim Rabba IJ, 12, 2). It is unclear, however, whether the
deviating translation of this section did not result (at least partially) from syntactical-
exegetical difficulties presented by some rather awkward Hebrew sentences.
154 CHAPTER TEN
original translator or a reviser). In the former case, the analysis of
Hebrew midrashic variants to MT is of interest for textual studies in
general, but does not bear on the study of the LXX, except for the
recognition that its Vorlage contained midrashic elements. However, it
is often very hard to decide whether a midrashic element, or any
exegetical deviation from MT, is based on a Hebrew variant reading or
tradition. As a result, the study of midrashic elements in the LXX has
its limitations,? even though such limitations were not taken into
serious consideration until the studies by D.W. Gooding.
Because of this difficulty, ideally a distinction should be made
between midrashic elements which were introduced by the Greek
translator, and midrashic elements contained in the Hebrew Vorlage of
the LXX. Such a distinction should be based on an analysis of the
translator's approach to his Vorlage, an analysis which is particularly
difficult in the case of Joshua. For, on the one hand this translation
contains many examples of very free exegesis in both small and large
details,? but it also reflects faithfully many details of its Vorlage,
inter alia many significant Hebrew variants.” The translation of Joshua
differs from many other books in the LXX in both the scope of free
exegesis and the large number of significant Hebrew variants reflected
in the translation. Hence, overall theories such as in the case of
Jeremiah on the one hand and Job on the other® cannot be applied to
Joshua. The analysis of midrashic elements in Joshua is thus more
complicated than the description of similar elements in Isaiah, Daniel,
and Proverbs and no clear distinction can be made between midrashic
elements presumably found in Hebrew manuscripts of Joshua and
Midrash-type exegesis introduced by the translator. Nevertheless some
examples of the latter type are adduced. The main purpose of this
3 Asa necessary result of this situation, many of the previously recognized ‘midrashic
elements’ or examples of ‘Jewish exegesis’ hardly belong to a discussion of the nature of the
sigs translation. This applies especially to many renderings analyzed by Prijs.
4 See especially the translation of chapter 6. For an analysis of the translation techniques
used in the LXX of Joshua, see J. Hollenberg, Der Charakter der alexandrinischen Uebersetzung
des Buches Josua and thr textkritischer Werth (Berlin 1876); S. Holmes, Joshua, The Hebrew and
Greek Texts (Cambridge 1914); and the dissertation of L. Mazor mentioned on p. 387. For
the text-critical problems of the LXX of Joshua, see H.M. Orlinsky, “The Hebrew Vorlage of
the Septuagint of the Book of Joshua,” VTSup 17 (1968) 187-195.
> See Tov, “Joshua.”*
© The great differences in length between the Hebrew and Greek texts of Jeremiah
derived mainly from the shortness of the Vorlage of the LXX (see Tov, "Jeremiah"*), while
the large omissions in the LXX of Job may be ascribed to its Greek translator: see Gerleman,
Job; D.H. Gard, The Exegetical Method of the Greek Translator of the Book of Job (JBL
Monograph Series 8; Philadelphia 1952); H.M. Orlinsky, HUCA 29 (1958) 229-271; 30 (1959)
153-167, 239-268.
MIDRASH EXEGESIS IN JOSHUA 155
short, non-exhaustive study is to illustrate some principles of Midrash-
type exegesis and to exemplify the methodological problems involved.
1. Midrash-type exegesis probably introduced by the translator
ZL MT two men
LXX — two youths (dvdpas in manuscripts MN... reflects a late
revision towards MT)
The ‘men’ (spies) have been explained here, as in the LXX of 2:23 (MT: two men -
LXX: the two youths) and 6:22 (see below), as ‘youths’ on the basis of 6:23 ‘the
spying youths,’ by way of gezerah savah: The spies are described in 6:23 as
‘youths,’ and hence they are to be depicted similarly in other places in the book. A
systematic attempt to make the description identical in Greek is found in 6:22, 23:
6:22 owint ow) -Kal tots Sualv veaviakots Tots KaTaoKoTE VoaoLv
6:23 orn oavin - ol &o veavloxot ol KaTackoTEvaavTes.
4:4 MT (Then Joshua called) the twelve men from the
children of Israel whom he had appointed (7737).
LXX _... twelve men of the distinguished ones (Tav Ev86Ewv)
of the children of Israel.
The translator described the twelve men who were to walk in front of the ark as
‘distinguished men.’ He may have referred to the ‘princes of the congregation’
(77¥N *NTwWI1), even though these are not called €vS0Eot. The princes of the
congregation are frequently mentioned in P as well as in Josh 9:15 ff.; 13:21; 17:4;
22:14 ff.” For a similar type of exegesis, see Exod 17:9 MT (And Moses said to
Joshua: ‘choose for us) men (and go out, fight with Amalek’) - dv8pas Suvatots (cf.
Exod 18:21 21 -wik ... TINN ANN) - Kal ob oeauTd oKéat ... dv8pas Suvatots).
4:5 MT (And let each of you take a stone and hoist it on his
shoulder) one for each of the tribes of Israel.
LXX _... according to the number of the twelve tribes of Israel
52 MT (At that time the Lord said to Joshua: ‘Make) flint
knives (0°08 ni29n) and circumcise the people of Israel
again the second time’).
LXX = pLaxalpas TreTplvas ék TéETpas AkpoTdLOU
(1) stone knives (2) of sharp stone®
” Thus MLL. Margolis, “taiv €v8é6Ewv —Josh 4:4,” Studies in Jewish Literature in Memory of
Prof. K. Kohler (Berlin 1913) 204-209. Margolis further mentioned two alternative
explanations of @v80€os: (1) the translator somehow explained ]?37 as ‘to honor’; (2) the
translator read 1)3 Wx, explained as ‘whom he honored,’ for which cf. Job 32:21.
Thus Rahlfs with manuscript B. Margolis, Joshua omits tetpl{vag with Aboxybz Arm
syh. However, this short reading probably represents a later revision towards MT even
156 CHAPTER TEN
The first part of the double translation reflects a literal rendering of MT, while
the second part is midrashic. According to de Lagarde, Proverbien,3, the free
rendering (2) originated with the original translator and was subsequently
revised and expanded with a literal translation (1). The midrashic rendering
explains the knives used by Joshua (0°) from the phrase Wx w*n?n, which occurs,
inter alia, in such a central place as Deut 32:13. dkpétojos renders w*n’?n in the
LXX of Deut 8:15 and Ps 113(114):8 as well as in ‘Theodotion’ in Job 28:9. The
midrashic element was inserted on the Greek level because it is reflected in the
Greek translation equivalent.
The same midrashic rendering recurs in the next verse (v. 3): paxyalpas
tetplvas dxpotépous. The literal translation occurs also in Josh 21:42d and
24:31a (Tas paxalpas tas teTplvas ), both in sections not found in MT.
a MT (And he circumcised the Israelites) at Gibat ha‘aralot
LXXMS5 w + kal €6nkev Oiwrlas akpoBuoTt(wv (and he formed
heaps of foreskins). = Lalué¢, Copt, and Just. Mart.,
Dial. Tryph., CXIII, 74-75.
This addition may reflect the OG since many unique elements in La and LXX!“ are
original (see Tov, “Lucian”*). The added words contain a midrashic explanation
of the name Gib&@t ha‘aralot (hill of the foreskins). According to the Midrash, this
hill was formed physically by a heap of foreskins, deposited there by Joshua.? The
midrashic explanation is reflected in several rabbinical sources such as Shir
Hashirim Rabba I, 12,3 (ay2. anem?7y 077 IVT), noted by M.L. Margolis and
D.W. Gooding.©
Seike) MT ~Op1 MRD... WIR
They ate ... massot and parched grain.
LXX Kal éddyooav ... d€vpa kal véa
They ate ... unleavened and new (corn).
"12 (parched) does not equal véa (new), and hence the following exegetical
procedure may be suggested. In Lev 2:14 2n73 wn) wR 777)2 2°38 is rendered by véa
Tedpuypéva xldpa épixtd. Because véa reflects 2°2x in this verse, as elsewhere in
the LXX (Exod 13:4; 23:15; 34:18; Deut 16:1), and medpvypéva reflects *19? (as in
though it omits the literal element in the rendering. In any event, the midrashic elements
are found in the whole Greek tradition.
According to some, however, the hill already existed while its name derived either
from the act of Joshua’s circumcision in loco or from his burying of the foreskins in that
particular place. Thus R. Gradwohl, “Der ‘Hiigel der Vorhaute’ (JJjosua V 3),” VT 26 (1976)
235-240.
10 Qohelet Rabba 11; Pirge de-Rabb1 Eliezer 29; Genesis Rabba 47:7 (on 17:23). Cf. M.L.
Margolis, AJSL 28 (1911) 5 and D.W. Gooding, “On the Use of the LXX for Dating Midrashic
Elements in the Targum,” JTS 25 (1974) 3.
MIDRASH EXEGESIS IN JOSHUA L57
Jer 29[36]:22 a’8’), this combination may have formed the basis for a new
equation, i.e., *17? - véa in Josh 5:11. A similar procedure was described by
Walters, Text, 179 with regard to the equation x*] = ToAvavdpetov in the LXX of
Ezekiel. Another possible explanation of véa is that it represents the rabbinic
term wn i.e., the new produce of the field which is not permitted to be used before
the Omer day (cf. Exod 23:10—14; m. Halah 1:1 and the dictionaries).
24:7 MT (And when they cried to the Lord, he put) darkness
(between you and the Egyptians).
LXX vedédnv kal yvdodov
...a cloud and darkness
The added element derives from Exod 14:20 which forms the basis for Josh 24:7
‘..coming between the host of Egypt and the host of Israel. And there was a cloud
and darkness (oxéto¢ kal yvddos ).’
The following ‘negative’ example is instructive from a
methodological point of view:
13:22 MT Balaam also, the son of Beor, the soothsayer, the
people of Israel killed with the sword, among the rest
of their slain (07°79n 5x).
LXX And Balaam, the son of Beor, the prophet, they slew
EV TH potr).
The meaning of form has been debated much (see, e.g., Schleusner, Thesaurus, s.v.).
LSJ explains the word as ‘victory,’ others as ‘struggle,’ while Frankel,
Vorstudien, 187 had explained the word previously as ‘throwing’ (in accordance
with the basic meaning of the root). Such a meaning may seem to be contextually
inappropriate, but Frankel explained the word as reflecting a tradition that the
Israelites killed Balaam by lifting him up and throwing him down. Such a
tradition is indeed reflected in TPS“) N to Num 31:8 x*nw MRI NAD P0177 _ NNN TV
MN) Wwe WINN. However, M.L. Margolis, JBL 33 (1914) 286-289 subse-
quently showed that the majority reading for was corrupted from a minority
reading tTpotn (aby*). The translator apparently explained the biblical 2%n as
Tpoty (rout), similar to 11:6 where o0°7°n is rendered by the related TetpoTw-
Lévous . He may have derived the word from ?1n/?"n, for which cf. 1 Chr 10:3 9n> -
étpoTuw6n.’! As a result, textual criticism (see Margolis for details) and
lexicography refute an otherwise laudable attempt to explain the translation of
13:22 as midrashic.
11 Cf. D. Weissert, “Alexandrian Analogical Word-Analysis and Septuagint Translation
Techniques,” Textus 8 (1974) 38.
158 CHAPTER TEN
2. Midrash-type exegesis either introduced by the translator or reflect-
ing Hebrew variants.
3:15 MT (The Jordan overflows all its banks) during all the
days of the harvest.
LXX _... as the days of wheat harvest (8eptopot trupaiv).
The additional trupaiv follows the word-pair o°un 1°xP or Bepiopod Tupav fre-
quently occurring in the Bible. Note a similar supralinear addition in 4QJosh? and
in the LXX of 2 Sam 21:10 Wp nonnn - Ev dpyij Septoyod KprOciv (barley harvest),
following the preceding verse (9: oMyw Wx).
4:6 MT For when your children (0371) ask you tomorrow ...
LXX For when your son (ulés) asks you tomorrow ...
The change from plural to singular is apparently based on Exod 13:14 = Deut 6:20
‘for when your son asks you tomorrow,’ both occurring in central sections. It is
not impossible that the change was influenced by the same phrase occurring in the
Haggadah, as in 24:4 discussed below.
The next verse (4:7) has been adapted in the LXX to the aforementioned change:
o7? onqnx1 (you shall tell them) - kal ob SnAdoets TH UL cov A ywv (and you
[singular] shall explain to your son saying ...).
5:6 MT For the Israelites traveled in the wilderness for forty
years.
LXX For forty-two years Israel wandered in the wilderness
Madbaritis (év Tj Eprjw tH] MadBape(T 81).
To the best of our knowledge, the tradition that the Israelites traveled in the
wilderness for forty-two years is not reflected elsewhere. It is hard to tell why
this deviating tradition is reflected in a verse which has no major importance for
the traditions about the wandering in the desert.
The forty-two years of the LXX, not a scribal error, somehow derived from the
typological number forty: in the Bible, the length of the period of the wanderings is
always indicated by the typological number forty (Num 30:13; Deut 2:7; 8:2, 4;
Amos 2:10; 5:25; Ps 95:10). In two instances, however, this number has been used
absolutely, forming the basis for subsequent calculations. In the present instance
the calculation is somehow related to Num 10:11. According to that verse, the
Israelites wandered two years before reaching Paran. Since they were punished
subsequently with forty years of wandering (Num 14:33 ff.), the whole period may
be calculated at forty-two years, as in the LXX of Josh 5:6. By similar reasoning
one could also deduct the two years from the total of forty. In this way a period of
thirty-eight years of wandering is created, as in Deut 2:14 (MT and versions):
‘And the time from our leaving Kadesh-Barnea (for which cf. Paran mentioned in
Num 10:11) until we crossed the brook Zered was thirty-eight years.’
MIDRASH EXEGESIS IN JOSHUA 159
5:10 MI They performed the Pesach offering (mo57 nx wy")
on the fourteenth day of the month in the evening.
LXX And the children of Israel kept (€tolnoav) the Pascha
(Td Tlacxa) on the fourteenth day of the month from
the evening.
In the Bible, no» denotes the ‘sacrifice of the passover’ and subsequently also the
‘festival’ itself (rarely). In postbiblical Hebrew, the second meaning prevailed. In
our verse, the biblical writer meant ‘the sacrifice,’ as may be inferred from the use
of 7wy which is a technical term denoting the performance of sacrifices. The
translator, however, understood no» as the festival, as may be inferred from the
end of the verse. For the sacrifice was slaughtered at the time of the evening, while
the festival was held from the time of the evening. The translator apparently
altered the meaning of his Vorlage by implying the second possibility. On the other
hand, if the translator actually read vn, no midrashic exegesis is involved, unless
one describes the very alteration of consonants as ‘midrashic’ (thus often Prijs,
Tradition, 54 ff.).
6:26 MT And Joshua laid an oath upon them at that time,
saying “Cursed before the Lord be the man that rises up
and rebuilds this city, Jericho. At the cost of his first-
born shall he lay its foundation and at the cost of his
youngest son shall he set up its gates.”
LXX ~ + And so did Hosan of Bethel; he laid its foundation in
Abiron his first-borm, and set up the gates of it in his
youngest son that had been saved.
The added words closely resemble 1 Kgs 16:34 but did not derive from that verse
in its present wording for they differ from both MT and the LXX in 1 Kings.!* The
added verse is probably based on a Hebrew source since S:aow0évtt apparently
reflects the root 21w} (the name aw is not reflected elsewhere in the Greek plus).
The history of the tradition of the rebuilding of Jericho may be reconstructed in
the following way. Originally neither the story in Joshua 6 nor that in 1 Kings 16
told of the actual rebuilding of Jericho. At a second stage a ‘deuteronomistic’ gloss
was appended to 1] Kings 16 telling of the fulfilment of Joshua’s prophecy. This
verse (34) forms a later addition since it is not connected to the surrounding
verses. This second stage is reflected in MT and all the versions except for LXX/"¢
(borcze2) which lacks the whole verse and hence attests the first stage (see Tov,
12 The differences between Joshua and Kings are as following in the B text:
Joshua kal év tu} €Xaxlotw Slaow8évtt ... Tag TUAGS a’Tis.
Kings kal Ta Zeyou8 TH vewTEpw abTod ... BUpas abnis.
13 Cf Prov 29:25, 21W° - owOjoeTat. For a detailed analysis, with different conclusions,
see L. Mazor, ”The Origin and Evolution of the Curse upon the Rebuilder of Jericho—A
Contribution of Textual Criticism to Biblical Historiography,” Textus 14 (1988) 1-26.
160 CHAPTER TEN
“Lucian”*). A third stage in the development is reflected in the LXX of Josh 6:26 to
which the contents of 1 Kgs 16:34 have been transposed. Although the additional
elements themselves are probably based on a Hebrew tradition, it cannot be
determined whether the Greek translator was responsible for its transfer to Josh
6:26 or the scribe of the Hebrew manuscript from which the translation was made.
7:1, 18, 19, 20, 24 MT ]aV
LXX Ayap (Axav in Axy*... is Hexaplaric)
The deviating transliteration occurs too frequently to warrant the assumption
that it was corrupted on either the Greek or Hebrew level. The transliteration
probably reflects an alternative spelling of the name of the main hero of chapter 7
which is closely connected with the aetiological explanation offered in v. 26
(mentioning the valley Dy) and embedded in v. 25 (‘n JIDy° WMD 7n). If the
etymological connection between Axap and 11Dy does not reflect the original
intention of the story, either it was part of the exegetical tradition surrounding
chapter 7 or it derived from the Hebrew manuscript from which the translation
was made (cf. 1 Chr 2:7 where the hero is called 13y). |
A similar problem arises in various places in the LXX (2 Kgs 21:18-25; 1 Chr
3:14; 2 Chr 33:20-25; Jer 1:2; 25:3; Zeph 1:1) where the name of Amon, the father of
Josiah, is spelled as Apws. In the LXX, Apes stands for both the prophet Amos and
Isaiah’s father Amotz,/* so that the transliteration of ]¥oX by Awe may refer to
either one of them. In view of the closeness of the names of Isaiah (17°yw>) and
Josiah (17°wx-), the Greek transliteration may refer to Isaiah’s father Amotz.
10:1 MI When Adoni-zedek, king of Jerusalem, heard ...
LXX When Adoni-bezek, king of Jerusalem, heard ...
A similar difference between MT and LXX occurs in v. 3. In several respects the
story told in Joshua 10 resembles Judges 1:5—7, although the differences between
the two stories are also manifest. For one thing, in MT the hero of the story is
called Adoni-bezek in Joshua but Adoni-zedek in Judges. In the LXX the
resemblance is stronger than in MT for in both places the king is called Adoni-
bezek. It cannot be determined whether the identification of the king in the LXX of
Joshua as Adoni-bezek derived from the translator’s exegesis or was found in a
Hebrew manuscript.
12:32 MIT ... from the children of Reuben and the children of Gad
LXX + and from half of the tribe of Manasseh.
Similar additions are found in vv. 33 and 34 in the same chapter. It seems that the
shorter textual tradition which omits ‘half of the tribe of Manasse’ and which is
14 Tn fact, the two persons are often confused as a result of the identical spelling of their
names, see especially L. Ginzberg, The Legends of the Jews, VI (Philadelphia 1959) 356-357.
For a detailed analysis, see S. Talmon and E. Tov, “A Commentary on the Text of Jeremiah,
I. The LXX of Jer. 1:1-7,” Textus 9 (1981) 1-15, esp. 7-8.
MIDRASH EXEGESIS IN JOSHUA 161
reflected also in Num 32: 2, 6, 25, 29, 31 (MT and LXX) reflects the original text of
these verses. The short formula has often been expanded in Hebrew manuscripts in
accordance with its full form, as happened in many similar instances. Such
expansion has taken place in the MT of Num 32:33 and in SP in that chapter (vv. 2,
6, 25, 29, 31). A similar phenomenon is reflected in the above-mentioned verses in
the LXX, but it cannot be determined whether the expansions derived from the
Greek translator or from his Hebrew Vorlage.
24:1 MT And Joshua assembled all the tribes of Israel to
Shechem.
LXX And Joshua assembled all the tribes of Israel to
Selo (Shilo).
24:25 MI And he drew up a Statute and an ordinance for them
in Shechem
LXX And he gave them a law and an ordinance in Selo
(Shilo) before the tabernacle of the God of Israel
The original reading in both verses is probably found in MT (Shechem) for the
terebinth mentioned in v. 26 connects the covenant with Shechem rather than Shilo
(cf. Gen 12:6; 35:4; Deut 11:30; Judg 9:6). Somewhat unexpectedly the central
covenant of the book of Joshua took place at Shechem which is mentioned
elsewhere in that book only in chapter 8 (and less significantly in 17:7; 20:7;
21:21; 24:32). This unusual situation probably accounts for the fact that the
location of the covenant has been altered to Shilo in the wake of such verses as
Josh 18:1 ff.; 21:2; 22:9, 12. For the importance of Shilo in the history of the
Israelite places of worship, see further m. Zebahim 14:6.
24:4 MT And Jacob and his sons went down to Egypt (alluding
to Deut 26:5).
LXX +and they became there a great, numerous and mighty
nation and the Egyptians afflicted them (Deut 26:5, 6).
The added words differ but little from Deut 26:5-6 ‘and he went down to Egypt ...
and they became a great, powerful and numerous nation °and the Egyptians
afflicted us.’ One element in the next verse in Joshua reflects the MT and LXX of
Deut 26:7 and not its counterpart in the MT of Joshua: Josh 24:7 MT ‘and they
cried unto the Lord’ - LXX ‘and we cried unto the Lord’ = Deut 26:7.
The addition to the LXX of v. 4 is not based on the LXX of Deut 26:5-6;!°
apparently it is based on a Hebrew tradition, but it cannot be determined whether
the translator himself turned to Deuteronomy or whether the addition was
already found in his Vorlage. It is not impossible that the translator or his Vorlage
was led to continue the quotation from Deuteronomy 26 beyond ‘and Jacob and
15 Josh péya kal Todt Kal Kpatardv = (MT: ovsy1 370 27772).
Deut péya Kal tArGos ToAb Kal péya (MT: 2 oy 3972).
162 CHAPTER TEN
his sons went down to Egypt’ under the influence of the tradition of the Haggadah
where the exposition of Deut 26:5-8 takes a central place. Although this
assumption is speculative, it should be pointed out that one possible influence of
the Haggadah on the LXX of Josh 4:6 was mentioned above. Furthermore, attention
should be paid to the evidence collected by Prijs, Tradition, 22 ff., 106-107 on the
similar interpretation in the Haggadah and the LXX of Deut 26:5 "28 73x mK, of
Deut 4:37; 26:8; Num 20:16; and Isa 63:8-9. On the connection between the latter
verse and the Haggadah, see also P. Winter.!®
24:12 MT (And I sent the hornet before you which drove them out
before you), the two kings of the Amorite.
LXX _... twelve kings of the Amorites
The number ‘twelve’ for the kings of the Amorites is not attested elsewhere.
However, even though a corruption of ‘two’ to ‘twelve’ comes easily to mind, the
assumption of an exegetical rendering or a variant ‘twelve kings’ is equally
possible. For the reading ‘two kings of the Amorite’ which is contextually
inappropriate!” may have been adapted to ‘twelve,’ possibly referring to the kings
of the west side of the Jordan who were subjugated by Joshua (the kings of Jericho,
Ai, the five kings of the Amorites [10:5] and the four kings of the North [11:1] are
together eleven). The variant ‘twenty-nine’ of LXX> refers to a similar list of
kings, viz. that of the kings listed in 12:24 (MT: thirty-one; LXX: twenty-nine).
24:15 MT we will serve the Lord
LXX + for he is holy
Cf. v. 19 you cannot serve the Lord for he is a holy God (MT and LXX).
24:32 MT (And the bones of Joseph which the Israelites brought
up from Egypt they buried in Shechem in the portion of
ground which Jacob bought from the sons of Hamor, the
father of Shechem, for a hundred pieces of 710°v;7)
Menlo. HOI ae:
LXX kal €8wxev abThv | wond év pepldi
.. and Joseph gave it in possession.
In MT the subject of 72n1? Or 7127 17>) is apparently ‘the bones.’ The Greek
translator, however, connected the second part of the sentence with the ‘portion of
ground’ in Shechem (iepls, note the female form of avtyv), when writing that
Joseph gave it in possession.!® It is not impossible that the interpretation of this
16 “Isaiah 63:9 (Gk) and the Passover Haggadah,” VT 4 (1954) 439-441.
17 The phrase ‘the two kings of the Amorite’ of MT is not appropriate because the text
refers to the west side of the Jordan river. However, this phrase may nevertheless be
original, if we consider the phrase to be an inappropriate (deuteronomistic) insertion (the
pane phrase recurs in Deut 3:8; 4:47; 21:4; Josh 2:10; 9:10).
8 Note the unusual equivalence 7?n1 - pepls in 32b which followed the equivalence
mn - pepls in 32a. The usual equivalent of 79n} is kAnpovoyla.
MIDRASH EXEGESIS IN JOSHUA 163
difficult verse was based on 05w in another difficult verse, i.e. Gen 48:22 (NX) 1X
oDv 7? "nm PnNX 2v. According to some sources (e.g., the LXX), this 05v denotes
the city Shechem while others explain the word as ‘shoulder’, ‘slope.’ In any
event, in both Gen 48:22 and Josh 24:32, the LXX refers to the ‘giving of (a portion
of ground in) Shechem.’
CHAPTER ELEVEN
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS IN
THE SEPTUAGINT
The three phenomena discussed here (loan-words, homophony and
transliterations) are similar inasmuch as all words involved resemble
their Hebrew counterparts and consequently they are often confused.
However, they are fundamentally different.
Thackeray, Grammar, 31-38 described the three phenomena as
‘Hebraisms in vocabulary,’ a term which should be limited to real
Hebraisms, that is, Greek words which are used in the LXX in an un-
Greek manner under the influence of the Hebrew language. He
distinguished between (a) ‘Hellenized Semitic words’ and (b) “Greek
words of similar sound to the Hebrew.’ Group (a) contains mainly loan-
words accepted into Greek from Semitic languages, but it also includes
Hellenized transliterations such as odBBatov and yetwpac. Group (b)
contains some “Greek words of similar sound to the Hebrew,’ but most of
the examples for this group are actually corrupted transliterations such
as dypov (from ayoup [Jer 8:7]), Ewe &Sou (from”O adwv [Jer 34(41):5]).
Walters, Text, 155-196, correcting Thackeray’s presentation, named
the general phenomenon ‘Hellenized Semitic words,’ a term which fits
only some of the words described by him. Walters began his discussion
by providing examples of words which should not be termed Hellenized
words, such as ‘corrupted Hellenizations.’ Besides these words,
Walters mentioned ‘groups for special treatment,’ such as ‘words
received into Greek more than once,’ ‘words borrowed from Aramaic and
not from Hebrew,’ and ‘borrowed words which took the form of already
existing Greek words of different meaning (homonyms).’ The latter
group, however, to which a long discussion is devoted (pp. 175-196), 1s
based on misconceptions not only in terminology but also with regard to
the phenomena described, as was pointed out in detail by J. Barr.!
Advancing the classification beyond Walters, G.B. Caird distin-
guished between (1) transliterations, (2) loan-words, (3) puns, (4)
1 VT 25 (1975) 247-254, esp. 249-250 (review of Walters, Text).
166 CHAPTER ELEVEN
natural or guided choice, and (5) mistranslations.2 Unfortunately, all
five categories together are named ‘homoeophony,’ and there is some
misconception with regard to the last three categories. In our view,
there is no room for a special category of puns (3). It is also unclear why
‘natural or guided choices’ (4) should be described as ‘simple phonetic
coincidence.’ Moreover, the characterization as ‘mistranslations’ for
such renderings as 027718 - avAlCwv (Jer 31(38):9), 72 - Bwyde (passim),
(8). - yf, (2 Chr 28:3; 33:6) is inappropriate (this group also contains
corrupted transliterations).
A detailed analysis and partial lists were provided by F. Knobloch
in a monograph on transliterations (1995) and by Lust (1998).3
Advancing the analysis beyond previous studies, we distinguish
between three different groups:
a. Loan-words, accepted into the Greek language from a Semitic
language.* This borrowing occurred in the time preceding the trans-
lation, and the words were subsequently used as natural equiv-alents of
their Hebrew counterparts.
b. Homophony (sound-resemblance), that is, the choice of Greek
equivalents which resemble the sound of their Hebrew-Aramaic
counterparts but differ in meaning.
c. Transliteration, that is, the transcription into Greek characters of
Hebrew and Aramaic words. Some of these transliterations were
corrupted in the course of their textual transmission to Greek words
similar in sound (see the appendix to section 3). The latter phenomenon
bears on the textual transmission of the LXX and not on the original
translation.
The phenomena described here illustrate different aspects of the
translators’ choice of equivalents at the level of linguistic exegesis,
sometimes involving a concern for the literary quality of the
translation.
As mentioned above, the phenomena are related inasmuch as all
three refer to Greek words which resemble the Hebrew words
represented by them. They differ from one another with regard to their
background because the resemblance to Hebrew words is derived in some
2 G.B. Caird, “Homoeophony in the Septuagint,” in: R. Hamerton-Kelly and R. Scroggs,
edd., Jews, Greeks and Christians, Essays in Honor of W.D. Davies (Leiden 1976) 74-88.
3B Knobloch, Hebrew Sounds in Greek Script: Transcriphions and Related Phenomena in the
Septuagint, with Special Focus on Genesis, unpubl. diss., University of Pennsylvania (Philadel-
phia 1995); J. Lust, “A Lexicon of the Three and the Transliterations in Ezekiel,” in:
Salvesen, Origen's Hexapla, 274-301.
Caird, “Homoeophony,” 79 (see n. 2 above) mentions also some loan-words from
Persian and Egyptian to both Hebrew and Greek.
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 167
cases from the vocabulary of the Greek language, in other cases from
the techniques used by the translators, and in again other cases from
textual corruption.
1. Loan-words
Already at an early stage, the Greek language absorbed several
Semitic words. Most scholars describe this phenomenon as borrowings
from Semitic languages, while others prefer to describe it as elements of
a common Indo-European-Semitic vocabulary. C.H. Gordon probably
went too far when describing Linear A as a Semitic language,° but in
Linear B Ventris and Chadwick as well as others recognized with
certainty some Semitic words.® Several Semitic words are also
evidenced in Homer. Usually scholars limit themselves to the
linguistic description of the vocabulary common to Semitic and Indo-
European languages, but some scholars suggested additional theories.
Thus C.H. Gordon and M.C. Astour stressed the common background of
Semitic and Greek culture in the second millennium BCE, especially as
evident in mythology. Similarly, according to J.P. Brown, ‘the common
vocabulary ... tends to generate parallel literary genres.’”
The theories of Gordon, Astour, and Brown are debatable, but the
fact remains that Mycenean sources contain such Semitic words as ku-ru-
so (xpuodéc = pn, or previously in Akkadian as hurasu or Ugaritic as
hrs), ki-to, plural ki-to-ne (xittTwv = nind in biblical Hebrew, or
previously in Akkadian as kiti, kutanu, or in Ugaritic as ktn(t)).°
These two words are also found in early literary sources starting from
Homer.
The evidence relating to words common to Greek and the Semitic
languages is described in etymological dictionaries and in several
° CH. Gordon, “Minoan Linear A,” JNES 17 (1958) 245-255; “Notes on Minoan Linear
A,” Antiguity 31 (1957) 124-130; Before the Bible, idem, The Common Background of Greek and
Hebrew Civilisations (New York 1962) 206 ff.; Evidence for the Minoan Language (Ventor 1966).
6 J. Chadwick, Documents in Mycenaean Greek (2nd ed.; Cambridge 1973) 135-136; M.C.
Astour, Hellenosemitica, An Ethnic and Cultural Study in West Semitic Impact on Mycenaean
Greece (Leiden 1967) 336 ff.; 5S. Levin, The Indo-European and Semitic Languages, An
Exploration of Structural Similarities Related to Accent, Chiefly in Greek, Sanskrit and Hebrew
(Albany 1971); idem, “The Accentual System of Hebrew, in Comparison with the Ancient
Indo-European Languages,” Proceedings of the Fifth Words Congress of Jewish Studies (1969),
IV (Jerusalem 1973) 71-77.
“Literary Contents of the Common Hebrew-Greek Vocabulary,” JSS 13 (1968) 163-191;
“The Mediterranean Vocabulary of the Vine,” VT 19 (1969) 146-170; “Peace Symbolism in
Ancient Military Vocabulary,” VT 2] (1971) 1-23.
8 See Chadwick, Documents, 135-136.
168 CHAPTER ELEVEN
monographs.’ It has been recognized that most loan-words can somehow
be brought under the common denominator of commerce, that is, words
which were easily transferred from one culture to another together
with the objects they denote. Among the loan-words, Lewy and Masson
(see n. 9) noticed groups of words referring to animals, plants, minerals,
clothes, food, coms and music.
While bilingual Hebrew/Aramaic-Greek inscriptions do not provide
evidence of loan-words juxtaposed with their natural Semitic
equivalents,!9 the LXX contains several instances of Semitic loan-words
used as natural equivalents of their Hebrew counterparts.
The view that a given Greek word is a Joan-word from a Semitic
source is often conjectural. Masson, Recherches (see n. 9), distinguishes
between different degrees of probability which may be ascribed to the
assumption that certain Greek words are loan-words from Semitic
SOUICES.
Below are listed some Greek words which had entered the Greek
language in the time preceding the LXX. These words are listed here as
equivalents of Hebrew words with the LXX, but several of these
equivalents existed already in earlier times between Akkadian and
Ugaritic on the one hand and Mycenean Greek on the other. Some such
loan-words are used as the only equivalents of a given Hebrew word:
TIN - woowmos (10 x)
27) - KGpndroc (54 x)
youn - xpuodc, tov (7)
nna? - (Bavog (19 x)
DO - GaTmeLpog (10 x)
MP w - OUKdULVOV (7 x)
7 E. Boisacq, Dictionnaire étymologique de la langue grecque (Heidelberg/Paris 1923); H.
Frisk, Griechisches etymologisches Worterbuch (Heidelberg 1953-1970); P. Chaqntraine,
Dictionnaire étymologique de la langue grecque (Paris 1968); A. Vanicek, Fremdwérter im
griechischen und laterimtschen (Leipzig 1878); K. Hartung, Septuaginta-Studien, Ein Beitrag zur
Gracitat dieser Bibeliibersetzung (Lamberg 1886) 13-18; E. Ries, Quae res et vocabula a gentibus
semiticis in graeciam pervenerint, quastiones selectae (Breslau 1890); W. Muss-Arnolt, “On
Semitic Words in Greek and Latin,” Transactions of the American Philosophical Association 23
(1892) 35-157; R. Lewy, Die semitischen Fremdworter im griechischen (Berlin 1895); M.-L.
Mayer, “Gli imprestiti semitici in greco,” Rendiconti dell’Instituto Lombardo, Cl. Lettere 94
(1960) 311-351; E. Masson, Recherches sur les plus anciens emprunts sémitiques en grec (Paris
1967). See also L.H. Feldman, JBL 96 (1977) 377 and Knobloch, Hebrew Sounds, 288-293 (see
n. 3 above).
0 Bilingual Greek-Phoenician/ Aramaic/Punic inscriptions have been collected by S.A.
Cook, A Text-Book of North-Semitic Inscriptions (Cambridge 1903); H. Donner and W. Rollig,
Kanaandaische und aramaische Inschriften \-II] (Wiesbaden 1964-1968).
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 169
Some of these loan-words are used as equivalents of Hebrew words
which occur very rarely in the Bible, some of them hapax legomena:
TIVDN - OPLG Job 20:16 (+ domtc¢ 1 x, BaotdkfoKog 1 x)
man - yadBdvn Exod 30:34
mow" - laotmic Ezek 28:13 (+ OovuxLov 3 x)
]02 - KULLLVOV Isa 28:25, 27
ae) - KpOKOG Cant 4:14
DDD ~ KApTaooc Esth 1:6
771 - vad pSoc Cant 1212; 4:13; 14
qn} - vi Tpov Jer 2:22
1°70 - OLVouv Judg 14:12, 13 (A), Prov 31:24
(+ 606vtov 1 x)
0D ; Ons Isa 51:8
75 - gakdc 1 Sam 10:1
1193? - KLVVGLLW|LOV Exod 30:23; Prov 7:17; Cant 4:14
MUN)? - Kaola Job 42:14; Ps 45(44), 8
Nw? - olkKuG Num 11:5
"on - TAWY 1 Kgs 10:22
In other cases, the loan-word is by far the most frequent equivalent of
its Hebrew counterpart:
gies) - KUVUpa (17 x) (+ kLBdpa 19 x, dpyavov 1 x,
wahktiHpiov 5 x)
nin> - XLTWV (25 x) (+ oToAH 1 x, Lluatvov 1 x)
W295 - Taddaky, KLE (36 x)(+ 'yurh 2 x)
172 - Képac (?) (92 x) (+ kedadt 1 x, cdATLyE 9 x)
TWWw - Tavpoc (17 x) (+ Bote 16 x, pdoxoc 43 x)
v2!) - OdKKOG (49 x) (+ dpotttrog 1 x)
DPw - olkhoc (35 x) (+ Spaxpr 1 x, otabude 1 x,
otTd@utov 3 x)
nn - TULLTIAVOV (?) (14 x) (+ avdde 1 x, badtHptov 1 x)
In again other instances, the loan-word is used sporadically:
155 - KUTIPOG Cant 1:14; 4:13 (+ 8 other equiv-
alents)
an OLBuc, OABn Exod 2:3, 5 (+ ktBwtée¢ 26 x)
As far as the frequency is concerned, the employment of Semitic loan-
words in the LXX reflects the general approach towards equivalents in
the LXX: Some equivalents were used sporadically, while others were
used passim in the LXX or in a single translation unit as the only or
major equivalent of a given Hebrew word or root (see Tov, “Dimen-
170 CHAPTER ELEVEN
sions”). The translators probably were aware of the naturalness of
these equivalents and of the similarity in sound. This awareness differs
from case to case since in some instances the Semitic loan-word was only
one of several equivalents used in the translation. Furthermore, the
resemblance in sound is often not easily recognizable. In the cases under
investigation the translators probably were influenced more by the
naturalness of the equivalence than by the resemblance in sound. It is
true that a certain tendency towards homophony may be recognized in
the translation (see group 2), but this tendency was the exception rather
than the rule. In light of this situation it is of interest to list some
presumed Semitic loan-words in Greek which have not been used in the
LXX as equivalents of their ‘natural’ counterparts:
detoc ~ oy
youu 7 shh)
Bdétratov, Bpdbu— - wrra
Epe Bog = ay
8UvVOG - ipaeg
Ka50cG - nis
KaVVa 7 13)?
KETPLG : NW?
KUL BaxXoG - yp /D
héEaxn : mow?
Nic - wr)
uLoyw - 117, ]0n
d8dvyj : ])OX
GovooVv - ww
oTupaé - “7X
There is no specific pattern in the distribution of Semitic loan-words in
the translation units of the LXX. In some cases one notes a
conglomeration of such loan-words, but this situation is determined by
the subject-matter. Thus Cant 4:14 mentions various scents and perfumes
which have been absorbed in the Greek language:
D-AwWI WRI ID OV 77787 V9 77129 -xy DD OV 719377711? ODID) T2
vdpSoc kal Kpdxog Kd\auoc Kal KivVdpiwpov peta TdvTwy EvAwY TOD
AtBdvov (AtBdvou?) opvpva ddw6 (dAon) WETA TAVTWY TPUTWVY LUpwY
2. Homophony
Several equivalents used in the LXX were chosen mainly or solely
because their sound resembled their Hebrew equivalents. In these cases
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 171
the translators preferred sound-resemblance (homophony) to identity
in meaning. This phenomenon is known from all translations, especially
from translations of literary texts, and hence, its occurrence in the LXX
is not surprising. Also in other aspects the LXX translation displays an
occasional concern for the literary quality of the translation,!!
especially in the poetical books.
Several examples of homophony have been recognized, often under
the wrong label,!* but caution should be applied. A condition for the
recognition of homophony is that the Hebrew and Greek words differ in
meaning. If the Hebrew and Greek words are identical in meaning or if
the Greek word reflects an exegetical rendering, the resemblance could
be coincidental. The analysis of similar-sounding words is often rather
complicated since equivalents may have been chosen because of more
than one reason. Some equivalents may reflect both linguistic exegesis
and an attempt to achieve sound-resemblance. For instance, the main
equivalent of 171, OAokavTwpa (an offering which is burnt fully)
represents the Hebrew well and, hence, the resemblance in sound
between 7719 and ddo- (cf. also another equivalent of 771y, ddoKdpTwpya)
may be coincidental. This pertains also to the main equivalent of 7570,
i.e. OqTptdAwToc (caught by wild animals).
Some examples of presumed homophony follow:
Ps 55(54):12 Rann FN Aan wy RX?
Kal otk €€€XTIEV EK TOV TAATELUY aUTIC TéKOG Kal &4d0¢6
Ps 72(71):14 ow5i 2x1 omni 71ND
éx TéKou Kal €E adikiac AuUTpWOETAL TAG WuXdG aUTOV
n, parallel to 107», means ‘oppression’ in general, while T6kog is used
in these verses as something more specific, ‘usury’ (cf. Lev 25:37 where
TOKog reflects qw31 and Deut 23:20 where éxtokl¢(w reflects the verb 4w3).
Hence, the translators of these verses (as well as of Jer 9:6) equated
1] See e.g, H.St.J. Thackeray, “The Poetry of the Greek Book of Proverbs,” JTS 13 (1912)
46-66; G. Gerleman, Studies in the Septuagint, II. Proverbs (LUA NEF 1:52, 3; Lund 1956) 11-
35; T. Muraoka, “Literary Device in the Septuagint,” Textus 8 (1973) 20-30; N. Leiter,
“Assimilation and Dissimilation Techniques in the LXX of the Book of Balaam,” Textus 12
(1985) 79-95.
12 See the studies by Thackeray, Walters, and Caird (n. 2) listed above, and further: P.M.
Ogerius, Graeca et latina lingua hebraizantes (Venetiis 1764); M. Heilprin, Bibelkritische
Noftizen, Ein nachgelassenes Manuskript (Baltumore 1893) 172-175; Wellhausen, Samuel, 10-11;
Driver, Samuel, 50 ff.; Barr, Comparative Philology, 58; Ch. Fritsch, “Homophony in the
Septuagint,” Proceedings of the Sixth World Congress of Jewish Studies (1973), I (Jerusalem
1977) 115-120; Knobloch, Hebrew Sounds, 308-311 (see n. 3 above).
172
CHAPTER ELEVEN
1(1)n with téKkoc because of the sound-resemblance, while the resulting
translation was contextually possible.
Ezek 21:36
Ezek 23:42
Nah 2:2(1)
mnwa -wan O72 Dw1X V2 7nnn (brutish)
Kal trapaSwow ae ele xetpac avdpuv BapBdpwv
TEKTALVOVTWY SLadbopdv (barbarous)
m2 12w yo7 D1 (multitude)
Kal Pwr appovlac avexpovovTo (music)
Jp Oy pp 79y (shatterer)
aveBN Eduadv el¢ todawtdév cou (one who blows)
Some of the cases of homophony involve inflected words, mainly verbal
forms. Such examples are more telling than the ones mentioned above:
Gen 42:1
Isa 24:19
Isa 35:6
Jer 6:29
Jer 31(38):9
Wann 77
Why do you look at each other?
tva Tl pabupet te
Why are you idle?
PAX WANNA Wp
The earth is rent asunder.
Kal dtopla dtropnéyjcetat Fh yr
And the earth shall be completely perplexed.
O>x pwd 79N7 (NOD 2°ND 177 TR)
And the tongue of the dumb shall rejoice.
Kal Tpavn €atat yiooa LoytAdAwv
And the tongue of the stammerers shall speak clearly.
NR OV
And the wicked are not removed.
Tovypla auTav ovK ETAKH
Their wickedness has not been melted.
on ~9n2 OX OD72IN OD IIN
I will lead them, I will make them walk by brooks of
water.
dvadEw avtovs avrAlfwy én Stupvyac Lédtwv
I will bring them (back), causing them to lodge by
channels of water.
On the other hand, if the LXX reflects a variant o1->x, no homophony should be
presupposed.
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS Navas
Since some of the renderings described as homophony refer to very rare
Hebrew words, even hapax legomena, they possibly do not reflect a
literary phenomenon, but rather a form of conjectural rendering (see
Tov, “Understand”*). In such cases the translators presumably
represented the Hebrew words with similar-sounding Greek words
because they found no better way of representing the difficult Hebrew
words. Some examples follow:
Isa 13:21 ON OF°N2 IX701 (OX OW 1X31)
.. Jackals
Kal €utrAno8yjoovtat al olklat AHyov
... sound, howling
mx occurs only here.
Jer 10:18 (PINT 7avY nx) yp un
Behold I sling.
LSov ya oxedlCw (Tobc KaToLKOiVTag Thy yr TavTnV)
Behold I upset.
v2? occurs only rarely in the Bible. The Greek word (‘to upset’) was probably
chosen because of the sound-resemblance. }4
1 Sam 13:21 72777 (273779)
.. the goad
Kal Ta Spemdvw Uidotactc qv h att}
... the sickle, curved sword
7247 occurs elsewhere only in Qoh 12:11. Cf. also v. 20 1nwinn - Kal 16 Spétravov
auTou.
In addition to homophony one finds in the LXX some cases of
onomatopoia, that is words which imitate the sound of the action
denoted.!4 If both the Hebrew word and the Greek word imitate this
sound, the two words are bound to be similar:
2)? - Btkoc Jer 19:10; cf. also 1 Kgs 14:3 MT
2? ; ddaddw (4 x in Jeremiah)
qn, 2/2 - hel yw passim
Pw - ouplCw passim
13 15] mentions only our verse as evidence for the occurence of this verb in the Greek
en epape: referring the reader to the compositum UrooxeN Cw.
4 Cf. A. Timmermans, Traité de l’onomatopée ou clef étymologique pour les racines
irréductibles (Paris 1890).
174 CHAPTER ELEVEN
The system of sound resemblance is also known for Aquila, who is known
for his literalness rather than concern for the literary quality of the
translation:!°
Thess: - avdwv (Deut 11:30)
710 - USwp (Lev 23:40; quoted in y. Sukk. 3:5 [53d])
wr : Ate VJJob 4:11)
zy a) - Kapxapoujevoc (2 Sam 6:16)
The phenomenon described here occurs only sporadically in the LXX
and is not characteristic of any of its translation units. The doubts
regarding many of the examples have been analyzed in detail by J.
Barr, “Doubts about Homoeophony in the Septuagint,” Textus 12 (1985)
1-77. The various examples of possible homophony have been
classified in Barr’s article according to the degree of their probability.
The phenomenon is distinct from transliteration (see section 3), but it is
understandable how Wutz reached his ‘transcription theory’ on the
basis of examples of both types.!®
3. Transliteration
A Hebrew or Aramaic word which was not translated but written in
Greek characters is considered a transliteration (transcription), such as:
Gen 35:16 9 M9x N19? PINT IID WV
evyéveto S€ hulka Hyytaev yaBpaéa elc yic €detv Edpaba
Three groups of Hebrew words were transliterated in the LXX:
a. proper nouns, B. technical terms, and y. unknown words.!” The nature
of these three groups differs, but they have in common that in all three
cases no Greek equivalent was used.
All transliterated words were corrupted in the course of their textual
transmission, see the additional note at the end of this study.
15 The examples are culled from Reider, Prolegomena, 153.
16 EX. Wutz, Die Transkriptionen von der LXX bis zu Hieronymus (BWAT IL, 9; 1925-1933).
According to this theory, the Greek translation was made from a Hebrew text which was
transcribed into Greek characters (similar to the second column of the Hexapla). The
examples of homophony and transliteration would then attest to that transliter-ated text.
However, most of the examples provided in support of the theory should be explained
otherwise, and, hence, the theory of Wutz cannot be supported by real evidence.
Bibliographical references to the reactions on the theory of Wutz are listed in Classified
Bibliography, 20 (‘Wutz’s theory’).
7 Most of the transliterations found in the LXX, culled from HR, are listed in N.
Simotas, Al duetddpactor AéEere ev Tuo KELpévw Tav O’ (Salonika 1969). See further Tov,
“Transliterations.”*
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 175
a. Proper nouns}®
Most proper nouns were transliterated in the LXX (including the name
of Pharaoh Necho in Jer 46(26):17 ty.nn TWayn Rw - Lawv EeoBt Euwnsd),
although occasionally they were translated:
Gen 31:48 9 7) yw NWP IDV... TY TIN 27
waptupet 6 Bouvdc ovToc ... Sta TOTO EkANSn Td Svoua
avtTot Bouvoc LapTupet
Gen 21:31 gw WwaAw) OY 7D YAW AND ...
Poéap dpk.opot Ste Exet Wpooav dugdteE por
Num 11:34 oO Nn Ova AX NP OW 7D NNT N77? ...
. Mvjpata tie ’EmO6uplac oti éxet €8awav Tov Aaov
Tov ETLOVENTHY
An accumulation of such translations of proper nouns is found in Deut
9:22, based on other verses (Num 11:3; Exod 17:7 etc.; Num 11:34):
Deut 9:22) 9 NN NPI WON WWAN) ...
Kal €v TH ‘Eptruptopss kal é€v Tad Tetpaopa kal év Tote
Mvypaow tij¢ ’ Emibuptac
Proper nouns are either single or compound, e.g., ‘house of ...”, “mountain
of ...’. As a rule, both constituents of the compound name were
transliterated, but sometimes their first element was translated.!?
Further research must determine which combinations were preferably
rendered by the former procedure and which by the latter.
Examples of the former are:
OX M2 - Baronr passim
DY pay - Epekaywp Josh 7:24
wp nwqwa - ev Larnpwd Kedpwv 2 Kgs 23:4
imv man. - é€v ApaBwé lepryw 2Kes 25:5
Examples of the latter are:?9
sgt US - 1 Baad pviav 2 Kgs 1:2, 3, 6, 16
mnwnian? - tov dpovc tot Moooad 2 Kgs 23:13
mw Ww - Hv TvUAny Tot loava Neh 3:6; sim. 12:39
m@p)0xn m3 - otkov Eoeduv 1 Chr 26:15; sim. v. 17
18 Cf N. Fernéndez Marcos, “Nombres propios y etimologias propulares en la
meL ap tay Sefarad 37 (1977) 239-259.
7 In again other cases both constituents were translated. See Fernandez Marcos,
“Nombres propios,” 249-250.
20 See further Fernandez Marcos, “Nombres propios,” 249.
176 CHAPTER ELEVEN
Due to differences between the translation units (and occasionally also
within such units), the employment of either a translation or
transliteration of a given name is inconsistent, as demonstrated by
frequently occurring names. Thus o°nw?® is transliterated as Pu\oTeLy
in the Hexateuch, but translated by d\\évaAo1 in the later books.2! 717%
was transliterated in Judg 13:8B; 16:28B and Ezekiel (passim), but
translated elsewhere by kvtptoc, 6ed¢, and S€oTé THe. 23 was trans-
literated in Joshua, Obadiah, Jeremiah b’, and Ezekiel a’, but
translated elsewhere by proc, Aly, peonyBpla, and vétoc.
Several common nouns were wrongly transliterated as proper nouns,
especially when occurring near other names or in the middle of lists:
Gen 28:19 (42¥x72 -PyT) ow n> O>IN?
kal OvAap Aout Av Svoya ...24
1Kgs 4:12 ovnp-? 732¥n IW
ews MaeBep Aovkay
Neh 7:7 (O8w> OV ~wIX) WON (TVA... YW? BAIT AY aKa)
Maodap
1 Chr 4:22) openy 097770 737) (2N? VI WR AW) WKY)
Kal dtTéoTpEewev avtouc aBeSnpiv aboukiv
1 Chr 8:7) 27-805 (8700 APN) JON)
oUTO¢ LyAaau
Most transliterations of proper nouns represent the phonetic value of
the Hebrew equivalents, as understood by the translators,23 usually
with the exclusion of the conjunctive waw and the prefixed prepositions
n?22. Occasionally suffixes were included in the transliteration:
Judg 13:2 B... "17 - Aavi (ard S4pLou auyyevelac Tod Aavi)
Judg 14:1 ff. nninn - @ayvaba (e.g., el¢ Bapvaba, Ev Bayvata)
Because most proper nouns in the LXX represent their Hebrew
counterparts, they are not declined. However, some names consistently
appear in their Hellenized forms such as Mwvorje¢ and are therefore
21 Fora possible explanation of the background of the variations and of the meaning of
ad\pvdog see R. de Vaux, “Les Philistins dans la Septante,” Festschrift J. Ziegler (Wurzburg
1972) 185-194.
22 Cf MG. Glenn, “The Word n° in Gen. xxviii: 19 in the LXX and in Midrash,” JOR 59
(1968) 73-75.
See the studies by Branno, Kénnecke, Lisowsky, Margolis, and Sperber listed in
Classified Bibliography, 18a, 20, and further Z. Ben-Hayyim, Studies in the Traditions of the
Hebrew Language (Madrid /Barcelona 1954).
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 177
declinable (see Thackeray, Grammar, 160-171).*4 The later biblical
books, such as Esther and the deutero-canonical books, contain a larger
number of Hellenized names than the early books. Josephus always
reflects the Hellenized forms of biblical names, while Jewish Greek
papyri from Egypt reflect both Semitic and Hellenized forms.”°
B. Technical terms
Several of the technical terms of the Torah were transliterated in the
LXX because no adequate renderings could be found. However, a given
word may be transliterated in one translation unit, but translated
elsewhere. Even within the translation units there is no consistency
with regard to the treatment of individual technical terms.
(1) Religion
At the time of the translation, the Jews of Egypt must have used in
their daily speech some untranslated Hebrew and Aramaic words
which denoted certain aspects of their religious life, and some of them
are now found in the LXX. Since Aramaic was the lingua franca in the
time of the translation, it causes no surprise to find in the LXX a few
transliterations which are based on Aramaic rather than Hebrew. See,
e.g., the transliterations of no) (Tacya), naw (caBBata),”° jn (Lavva), 71
(yeLwpag = 81171). Likewise, many transliterated Hebrew words ended in
-etv rather than -e.y (see Seeligmann, Isaiah, 65, n. 40), possibly under
the influence of rabbinic Hebrew.
Most of the transliterated technical terms in the LXX belong to the
realm of religion in the wide sense of the word:
OPN - atAau, ovAan (also translated apnttc, vadc)
TION - edoud, egw (also translated €mwulc, oToAH)
nn - Baya (also translated d\ooc, Bouvéc, Bwydc, et Swrov,
Epnyov, PBuataoTiptov, oTHAAN, bdmAdc, tysoc)
pkeoie - d5aBetp (also translated vadc)
mianta - uavaa (also translated Supov, 6vola)
Baia - XEPOUBELV, “ELL
orn - vabeLvipy
24 See further C. Krénnecke, Die Behandlung de hebrdischen Namen in der Septuaginta
(Programm des Koeniglichen und Groening’schen Gymnasiums zu Stargard in Pommem;
Pipa 1885) 27 ff.; Helbing, Grammatik, 58-60.
29 See N.G. Cohen, “Jewish Names as Cultural Indicators in Antiquity,” JS] 7 (1976) 97-
128.
26 See A. Pelletier, “SaSBara, transcription grecque de l’araméen,” VT 22 (1972) 436-447.
178 CHAPTER ELEVEN
nop 7 TAaCXYa
oonw - oapadetv
o-pnn - Qepaduy (also: etSwAra, kKevotddia, yAuTTG, ShAoL)
naw - oaBBata (cf. n. 26)
(2) Measures, weights
D°N : ott (also translated étpov, Téa, TAAaVTOV)
Ta - €LV
ann - yo\.op (also translated dptdBn, kdpoc)
ae) - 6 KOpOG
mya)¥) - -yonop (also translated Spdyya, pupdc)
Most transliterations of technical terms represent the phonetic value of
the Hebrew words as understood by the translators, including the
Hebrew morpheme of the plural, e.g. Baadtp (or sim.), xepouBetp (or
sim.), we xwyw0d.27
Some transliterations add Greek suffixes:
na - 6 Bd60cG, 6 Batog (Ezra 7:22; also translated kapdiov,
KOTUAT LETPOV, WETPTITNG, YOtvLE)
‘ae : ol yacapnvol (Dan 2:27; 5:7, 8)
72 - 6 KOpoG (passim)
V1) - dvactipatog (Judg 13:5, 7; 16:17, all in manuscripts
A...)
1? - 6kdBog (2 Kgs 6:25; hapax legomenon)
It is difficult to understand why only certain transliterations received
Greek suffixes. In some cases the existence of Greek homonyms must
have been influential, such as the measures kdpog (cf. 6 Kdépoc ‘satiety’)
and Bdtocg (cf. } Batoc ‘bramble’).
The Hebrew text of certain books contain more technical terms than
others, so that the frequency of the transliterations in the LXX cannot
be compared. At the same time, the preferences of some translators can
be recognized because some Hebrew words were transliterated in certain
books, but translated in others (see above). One thus notes the frequent
transliteration of technical terms in 2 Kings (see below).
27 Transliterated technical terms usually behave according to the rules of the Greek
language with regard to the article and number (e.g., T6 alAdp, TO Lava, Ta xepovBetpy, Tuy
XepouBett, TAG Laxywvwd). However, under the influence of the Greek context one also
finds such forms as f pexwvwd (1 Kgs 7:30 (17), 34(20); this forma mixta was created by the
singular form in MT and the plural pexwvw8 occurring passim in the chapter) and 16
Sepadiv (Judg 18:17, 18, 20; cf. the parallel nouns with 16 in the Greek contexts).
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS 79
(y) Unknown words
Since there is no evidence for the existence of dictionaries nor word-
lists, translators had to resort to different sources of lexical information
such as exegetical traditions, the context, etymology, postbiblical
Hebrew, Aramaic and for the later translators also the translation of
the Torah (see Tov, “Pentateuch”*). When these sources were of no
avail, translators turned to conjectural translation (see Tov, ”“Under-
stand”*), and in more extreme cases they left the word altogether
untranslated.
When words were left untranslated, the translator may have hoped
that the transliterations would be replaced by translations. The
transmitted evidence gives us some reason to believe that such trans-
lations were indeed added for in several cases an original translit-
eration is transmitted together with a subsequent translation:
Gen 22:13 702 - Ev buTw aaBeK
1Sam5:4 9 ynann ON - éml Ta €utTrpdabta apaded
1Sam 6:15 TNT NN) - Kal TO Géua epya® (cf. vv. 8, 11)
2Kgs 10:22 aAnnenn - Tov olkou HeaOaad
Often it is unclear whether a word was transliterated because it was
unknown to the translator or because of other reasons. However, it
stands to reason that the former was the case when the Hebrew word is
either a hapax legomenon or very rare, as in the following cases (for a
complete list, see Tov, “Transliterations”*):
Gen 36:24 (1272) On°7 (NX Xn) - TOV Tapitv
Judg 8:7 B (cf. v. 16) 0732927 nx) (N27NT OR? NX ODWwa] nx "nw7)) - (Ev...)
Kal €v tate aBapknve.v
1 Kgs 5:25(11) 1°39 nD (Bron AD PX a wy) - Kal paxtp TA olkw alto
2 Kgs 8:15 12907 Np - kal €AaBev TO Laypa
1 Chr 21:20 O-xanna wy VID Nya) - Kal Téooapes vLol avUToD WET avTod
ePayaBLv
The largest number of transliterations of unknown words occurs in 2
Kings and outside the LXX in the sections ascribed to kaige-Th. Since
the ‘LXX’ of 2 Kings actually belongs to this revision as well (see
Barthélemy, Devanciers), it was suggested by Tov, “Transliterations”*
that this reviser preferred to transliterate unknown words rather than
to make guesses at their meaning.
As a rule, unknown words were transliterated in their exact Hebrew
form, including prefixes and suffixes, e.g.
180 CHAPTER ELEVEN
Judg 5:22 A... pT aX NNT N17 - awadapwO Suvatay abtod
Ezek 41:8 2°20 3°20 79) "27 "YN - Kal Td Opaed Tod olkou tbo
KUKAW)28
Ezra 8:27 3n1 “5D) - kal Kadoupn ypucot
1 Chr 28:11 von (nr mi... 1971) - kal THV Cakxw abTtod2? (cf. v. 20
LXX)
Several common nouns have been transliterated in the LXX as nouns
probably because they were unknown. This refers in particular to rare
words, some of which are hapax legomena, and contextually difficult
words:
Gen 15:2 -m-2 pwn 73) (hapax legomenon) - 6 8€ vldg Macek THe
olkoyevotic® Lov
Gen 35:16 AN7DX N12? PANT 0722 Ty 77m - Eyéveto S€ Hwika Tyytoev
xaBpa@a ele yriv édMetv Edpaba
Gen 48:7 ANTDN XI? PIN 72D WWI W772 - EyyiCovtdc tov kata TOV
imm@dé8popov xaBpaba?! nig yiic Tob édOetv Eppada
2 Kgs 5:19 pax 1799 1nxNn 72 - Kal dmyAGev at’ abtot elc AeBpaba tic
yiic
1 Sam 20:20,79009 79 now) (rare word) - éxnéuttwv el¢ THY Appattapet
1 Sam 24:2 (a-)y> rare word) O°2977 7778 71D Oy (PwWIRT TIT DN WPI? 72>)
-étl tredowtov Za(S)Sarep
1 Kgs 15:22 97) XR ATW 72 DR NWT NOX JNM - Kal 6 Baothede Aca
TapnyyelvAev travTl lovda ele ALvaKLUL
Ps 74(73):15 jr°x nani (nwain nx) - Totapove Hbay
28 Corrupted transliteration of the Hebrew. Probably the lamed of the next word is
included in the transliteration.
The transliteration probably reflects 31 (the Hebrew pronominal suffix is represented
twice) while 1n (531) is represented by kal.
The Greek probably represents a contextual rendering of *n-2 rather than a second
rendering of ;wn.
Double rendering of n725.
LOAN-WORDS, HOMOPHONY, AND TRANSLITERATIONS
Lod
Additional note: Transliterations corrupted to simtlar-looking Greek
words
Transliterations of Hebrew words were subject to corruption because
they were not understood by Greek scribes, as exemplified here:
Judg 5:16 onown
AMNab’ ckmy uooda.bapy
On woode Bay
bdptvwb2 Arm woodabaty
n Thdt-ed woodabeu
g LLoogavev
] Loc dbe
Thdt-cod noopabe
Dia SuYOULaG
1 Kgs 5:25(1) n?2n
hi waaxLr
AMNf£gj™tmq-z Uayad
n waaxar
La machit
d [LaxXaTL
P UayaTh
Eth maacheto
e xar
rel ULaYeLp
As part of the process of corruption of the transliterations, some scribes
tended to alter such transliterations into similar-looking Greek words
even if the resulting Greek word was inappropriate:
1 Kgs 18:32 (a°nxd n°2D) 72yn (wy)
Oc7*ep Qaaka
bgz(mg)c4? 6adhaa
dip Q@adaav
rel édAacoay?2
1 Sam 15:23 7857 G|TN) WN) 7 (OOP NXvM 7D)
Bvy depatrelav
1 B€ PATELV
] Be padnvy
cm Cepadip
x GEpadiu
ANaz(txt) rel Be padetv
32 See Walters, Text, 190-192 for a different view.
182 CHAPTER ELEVEN
The original version of 1 Sam 15:23 probably contained a transliteration
of o-5"m, as often elsewhere in the LXX, while Sepatretav is secondary.°*?
In some cases the corruption must have been very early because it is
contained in all extant manuscripts. In such cases the original
transliteration must be reconstructed by way of conjecture:
Jer 31(38):21 o-nnnn 72 cw] All manuscripts Tol{noov Tiuwplav;
Spohn : tipwpip.
33 See Walters, Text, 180-182 for a different view.
34 M.G.L. Spohn, leremitas vates e versione 1udaeorum alexandrinorum (Lipsiae 1794-1824).
CHAPTER TWELVE
THE IMPACT OF THE SEPTUAGINT TRANSLATION OF THE
TORAH ON THE TRANSLATION OF THE OTHER BOOKS
According to ancient evidence as well as modern descriptions, the
translation of the Torah preceded that of the later books. As might be
expected, this translation had an impact on the translations of the
other biblical books which were made subsequently.!
From the outset it was only natural that the Greek Torah would
influence the translation of the subsequent books. The reading of the
Torah in the synagogue started at an early date,* and although there is
no sound evidence for the liturgical use of the Greek Torah nor of the
later Greek books,’ the Torah must also have been widely known in
Greek. In the following discussion, evidence is presented in support of
the following four points:
1. The vocabulary of the Greek Torah was continued in the translation
of the later books.
2. The Greek Torah served as a lexicon for the later translators who
often turned to that translation when encountering difficult Hebrew
words.
3. Quotations from and allusions to passages in the Torah in the later
books were often phrased in Greek in a manner identical with the
translation of the Torah.
4. The contents of the Greek Torah often influenced the wording of
later translations on an exegetical level.
1 Thus H.St.J. Thackeray, “The Greek Translators of the Prophetical Books,” JTS 4 (1903)
583; M. Flashar, “Exegetische Studien zum Septuagintapsalter,” ZAW 32 (1912) 183-189; A.
Kaminka, Studien zur Septuaginta an der Hand der zwolf kleinen Prophetenbicher (Frankfurt a.
M. 1928) 17-20; J. Ziegler, Untersuchungen zur Septuaginta des Buches Isaias (ATA XII, 3;
Miinster i. W. 1934) 134-175; G. Gerleman, Studies in the Septuagint, II, Chronicles (LUA I, 43,
3; Lund 1946) 22 ff.; Seeligmann, Isaiah, 45-49; Allen, Chronicles, 23-26, 57-59.
27. Elbogen, Der jiidische Gottesdienst in seiner geschichtlichen Entwicklung (3d ed.; Frankfurt
1931) 155ff. (see also the updated Hebrew translation of this book [Tel Aviv 1972]).
3 Pace HStJ. Thackeray, The Septuagint and Jewish Worship (Schweich Lectures 1920;
London 1921).
184 CHAPTER TWELVE
1. Vocabulary
The translators of the Torah created a translation vocabulary of
Hebrew-Greek equivalents, the foundations of which were probably
laid in the generations which preceded that translation.* The nature of
this vocabulary must be the subject of future research,” but it is safe to
say that one of its main characteristics is the lack of variation, a lack
caused by the translators’ frequent use of fixed equivalents.
When investigating agreements between the vocabulary of the
Greek Torah and that of the later books, obvious agreements such as 7UX
- yur7], WX - dvOpwro¢g, wow - hALog and y77Dx& - Bdtpaxos, are disregar-
ded. The analysis concentrates on agreements which are idiosyncratic,
and illustrates how the translation vocabulary of the Torah was
continued in the translations of the other units. In the comparison of the
vocabulary of the books of the LXX the following points are taken into
consideration:
1. The degree of dependence of the later books on the vocabulary of
the Torah cannot be expressed in absolute statistical terms because
each book was rendered by a different translator. Nevertheless, on
the basis of pilot investigations one can describe the vocabulary of
certain books as more ‘Pentateuchal’ than other books. For example,
Gerleman, Chronicles (see n. 1) described the vocabulary of Chron-
icles as more ‘Pentateuchal’ than that of the parallel translations
of Samuel-Kings. See further n. 9.
2. All books (or sometimes sections of books) were rendered by
different individuals and necessarily differed in character and
translation vocabulary,® although all translators adhered to some
extent to the vocabulary of the Torah.’ One is therefore justified in
investigating the influence of this vocabulary on that of the later
translations.
4 See my study “Studies in the Vocabulary of the Septuagint,” Tarbiz 47 (1978) 120-138,
esp. 137-138 (Heb. with English summary).
° Fora partial study, see Daniel, Recherches.
© These differences justify our neglecting differences between the individual trans-
lations. Little can be learned from disagreements in vocabulary between different
translators (even in whole verses or sections which are identical in the Hebrew Bible)
except for the lack of cooperation among the translators and their failure to consult other
translation units. For a different approach, see C. Egli, “Zur Kritik der Septuaginta. Sind die
Hermeneuten des Pentateuch und des Buches Josua identisch?” ZWT 5 (1862) 76-96, 287-
321; A. Kaminka, Studien, 17 (see n. 1 above).
7 See the article quoted in n. 4 and further Tov, “Dimensions.”*
IMPACT OF THE TORAH 185
In the examination of the dependence of the later translations on the
vocabulary of the Torah the examples are subdivided into:
a. religion
b. legal terminology
c. central biblical terms
d. miscellaneous.
The non-exhaustive list below concentrates on equivalents occurring
several times in the Torah (see further the list in Tov, “Samuel,”* 338—
339). The following conventions are used in the List:
For two or more occurrences in one book one reference with ‘etc.’ is
given.
For two occurrences in two different books precise references are
given.
If there are more occurrences in different books p. = passim is used.
Words denoted with an asterisk (*) were presumably coined by the
translators of the LXX (‘neologisms’).8 Several of the neologisms of the
Greek Pentateuch were continued in the later books, a situation which
underlines the dependence of the latter on the former. The inclusion in
the list of certain equivalents does not imply that the Greek word
mentioned is the only equivalent used for the Hebrew one. Lust,
“Vocabulary” offered additional examples for the LXX of Ezekiel
(especially Leviticus 26 compared with Ezekiel) and he also noted that
this translator is less ‘Pentateuchal’ than may be expected.?
a. Religion
a. General
adyxtoTe vo 2X1 p p
ay.ac- an ,71 Lev 25:11;Num 6:12 p
dns IND p Pp
aKkaapola MND p Pp
axabaptés NDU Pp 2)
avdabrua (-e1a) on p Pp
amapxy) YUN p Pp
WMMN p )
8 The word under consideration was probably coined by the translators (or a preceding
generation) in order to express Hebrew words and concepts which, in their view, could
not be expressed adequately by existing Greek words. Our observations are based on the
evidence listed by LSJ and LSJ, Supplement. The assumption of a ‘neologism’ is subject to the
limitations and doubts such as described in Tov, “Compound Words.”*
Lust also pointed out that Pap. 967 often goes its own way, away from the vocabulary
of the main manuscripts of Ezechiel, which may have been revised in these cases.
186 CHAPTER TWELVE
adalpepa*lO Wann p Ezek 44:30
yuTTTOv 20D p Pp
SexTds * pxv p p
StraéyKn mi p P
Su5pov onmin Gen 4:4 etc. Pp
a? =p Neh 13:31
nw p p
elSwdov o°7171 Lev 26:30; Deut p
29:16
EKOUOLOV* nada Pp
éEtAdoKouar pig 28a © Pp
€ESSLOV Mxy p 2 Chr 7:9; Neh 8:18
émLS€KATOV Wwyn p Pp
émeTlOnpe (a°7) JaO p 2 Chr 29:23
YT op 2 Kgs 5:11
éEtupls TIDN p 1 Sam 21:9(10)
QOvoLaoTTpLov* nan p Pp
KLBwtos JR p p
MELTOUP YEW mu p Pp
VOLLoS mn p Pp
dou evw&las niin y |p Ezek 6:13 etc.
TAN WE AELa OowX p Pp
TIPOOKUVE W) mnnwit p p
TpwToyevvr|ata* on22 p 2 Kgs 4:42; Neh 10:36
TpwTdToKOS * Nod 'p Pp
OTHAN max p Pp
ouvayuryn} n7yv p p
OWTHpLov ow p p
yetpotrolntos 2 Ox Lev 26:1 Isa 2:18, etc.
xwveuTtds * moon op Pp
Furthermore, the translator of Isaiah accepted from the translators of
the Torah the distinction between the Jewish natn (@votaoTpLov) and
the pagan natn (Bwpds).!!
B. Technical terms
The post-Pentateuchal occurrences of the words are not necessarily
found in religious contexts, e.g. kpaTHp in Cant 7:3(2).
10 Lust, “Vocabulary,” 534 points out that Pap. 967 of Ezechiel uses d@dépioya instead
of dpalpeyia, so that the equivalent of the main manuscript tradition of the LXX may
have been secondary.
11 For an analysis, see Daniel, Recherches, 18-19.
IMPACT OF THE TORAH 187
ywvla yypn Exod 26:23 ff. 2 Chr 26:9; Neh 3:19 ff.
S5éppis m1 Exod 7 ff.;. Num 4:25 Pp
SLyoTéunLa mn Exod 29:17; Lev 1:8 Ezek 24:4
*
BulaKn* Dm p Pp
Ouplaya mvp p Pp
kaTamé Tao mw» p 2 Chr 3:14
iat
KlSapis mixn Exod 28:4 etc.; Lev 16:4 Ezek 21:31(26)
KpaTtr|p xX Exod 24:6 Cant 7:3(2)
Kped ya 1910 Exod 27:3 etc. ; Num 4:14 Pp
aBls onj2n Exod 37:23(38:17); Num 4:9 Isa 6:6; 2 Chr 4:21
\d-yavov PPT p 1 Chr 23:29
héEBNS yo Exod 16:3 Pp
houTtp ava «Exod 30:18 etc. ; Lev 8:11 Pp
huxvla aaa: -p Pp
aXe nny Exod 29:17; Lev 1:6 etc. Judg 19:29B; Ezek 24:6
WLETPOV mx Deut 25:14, 15 Pp
unpla 70> Lev 3:4 etc. Job 15:27
Lupe pds m1 Exod 30:25 etc.
Té TAAOV y°x Exod 28:36 etc.; Lev 8:9 1 Kgs 6:18 ff.
TANPWOLS ox?» Exod 35:27 1 Chr 29:2
TUy|LT| AMX Exod 21:18 Isa 58:4
Tupetov mn p 2 Kgs 25:15; 2 Chr
4:21(22)
poloKxos* }97 «Exod 28:33(29) etc. 2 Chr 3:16 etc.
odpS.ov o1x Exod 28:17 etc. Ezek 28:13
oeplSadts neo p p
ouvpva ™ Exod 30:23 P
aotlyos 0 Exod 28:17 etc. 1 Kgs 6:36 etc.
OUvOECLS no =p pe Sevag sa
THyavov ronn Lev 2:5 etc. Ezek 4:3; 1; Chr 23:29
TopeuTéos mn Exod 25:18(17) etc. Jer 10:5
b. Legal terminology
The post-Pentateuchal occurrences of the words are not necessarily
found in legal contexts.
(Td) BLBALOV (Tod) ~o Deut 24:1, 3 Isa 50:1 Jer 3:8
dtootaalov mm7D
188 CHAPTER TWELVE
auTdx6buv nx p Josh 8:33(9:2)
Ezek 47:22
YELWpAS ay Exod 12:19 Isa 14:1
yovoppur|s* a1 Lev 15:4 etc.; Num 5:2 Pp
éyyaaTpluos ax Lev 19:31 etc.; )
Deut 18:11
éve xupdcw ?an p 9)
étraoidds * U7? Lev 19:31 etc. 2 Chr 33:6
6@vno.atov* 7221 Lev 5:2 etc.; Deut 14:8ff. )
KAndov( Coa” jay Deut 18:10 2 Kgs 21:6; 2 Chr 33:6
\étTIpa + deriva- ny1x Lev 13-14; Deut 24:8 Pp
tives
ALBoBorAE w* 20 p p
n”10 p Pp
uavrela, Od}? p p
\LavTetov
LOX LOS nin p °)
Tpdos ay) Lev 25:14 ff.; Deut 18:8 Neh 13:20
ouvTluinoLs Ty Lev 27:4 ff.; Num 18:16 2 Kgs 12:5 etc.
dapwakds * mwxn) Exod 7:11 etc.; Pp
Deut 18:10
gpuyadeuTApLov* vpn Num 35:6 ff. Josh 20:2 etc.; 1 Chr 6:52
ff.
Appendix: Clean and unclean animals
Sopkds *28 ~=Deut 12:15 etc. Pp
éX\ados (7)9°x Deut 12:15 ete. P
TBis mwr Lev 11:17 Isa 34:11
Lis “aay ~=Lev 11:29 1 Sam 6:4 ff.; Isa 66:17
vukTepls mov Lev 11:19; Deut 14:18 Isa 2:20
vuKTLKdpaé o> «=Lev 11:17 Ps 102(101):7
XOLPOYPUAALOS * jpw Deut 14:7 Ps 104(103):18;
Prov 30:26 (24:61)
c. Central biblical terms
Many of the words in the Torah pertain to central stories or issues to
which reference is made in the later books (note especially Psalm
78[77]). The later translators often used the same Greek equivalents as
were used in the Torah:
Stayoyylfw*
51é Ea8Sos
Ta EptTEeTA
KATAKAVOLOS
KUVOLLULa
Lavva*
OpTVYOLNTpa
TapdSe.aos
TAGE
OTEpEWLLA
davtats *
] Ponti
ney
d. Miscellaneous
a. Technical terms
dvepopbopla*
BovTupov
S{8paxpov
So0K6s$
éykpudlas
éSéopata
EVOTLOV
émaoLdé¢
ETMAVALS
ETLOLTLOILOS
Ad Xava
uavdpaydpas
LLOALB(S)os
LLOVOKE pws
LLoAuKY)
vuLoaywydos
dBords
ol dt*
dv\Upa
Oust Los
TayeTos
pow
nnod
wip2n
ny?
IMPACT OF THE TORAH
Num 34:4 ff.
Gen 6:17 etc.
Exod 8:17(21) etc.
Josh 9:18
Josh 15:4 etc.; Ps
68(67):20
P
Ps 29(28):10
Ps 78(77):45;
105(104):31
189
Pp Ps 78(77):24
Exod 16:13; Num 11:31, 32 Ps 105(104):40
Gen 2:8 etc. Pp
Exod 31:18 etc.; Deut 4:13 etc. 1 Kgs 8:9; 2 Chr 5:10
Gen 1:6 ff. p
Gen 1:15 Ps 74(73):16
Deut 28:22 2 Chr 6:28
Gen 18:8 Deut 32:14 P
Pp Josh 7:21; Neh 5:15 ete.
Gen 19:8 2 Kgs 6:2, 5; Can 1:17
Pp P
Gen 27:4 etc. Prov 23:3
Gen 24:2? etc.; p
Exod 32:2 etc.
Exod 7:11 etc. Dan 2 2727
Gen 25:16; Num 31:10 Ps 69(68):26
Gen 42:25 etc.; Pp
Exod 12:39
Gen 9:3 Ps 37(36):2
Gen 30:14 ff. Cant 7:14(13)
Exod 15:10 p
Num 23:22; Deut 33:17. Job 39:9; Ps 22(21):22 etc.
Gen 4:23; Exod 21:25
Gen 26:26
P
Lev 5:11 etc.;
Num 5:15 ete.
Exod 9:32
Deut 11:14
Gen 31:40
Isa 1:6 etc.; Ps 38(37):6
Judg 14:20
Ezek 45:12
P
Ezek 4:9
P
Jer 36(43):30
190
TE PLBdAAaLov
TE pl xwWPOS
TIUppaKns
otraptlov
TLONVdéS
Tpobds
Tpvyuv
Dooulo¢g
dakds
xelp
Xoxn
xXopTao\.a
xvUTpa
Wat) pLov
BedXNLOv
Udyos
B.
yéveats (-ets)
YPayLLate bs
Bf\Los
duyynia
Slppos
EKAEKTOS
KATAOXEOLS
LLaKkpd@uos *
.dx80¢
OAKT)
dp8plCu*
dsapus
TAAALOW
TapatiKkpalvw *
Tove AE OS *
ToTaLd¢g
NOD
MWY
jan
WIT DRI
NIDON
T17KR)D
WD
TAX
man
7Dw
mown
mw
NOD
N71
TInN
O7DN JN
mon
Wn
o-Dwn
o-x>n
Tee
m7
Jon 37
qN?
CHAPTER TWELVE
Exod 22:26; Deut
2212
Gen 13:10 etc.
Gen 25:25
Gen 14:23
Num 11:12
Gen 35:8
si
P
Gen 25:34
Exod 9:8; Lev 16:12
Deut 29:17 etc.
Gen 24:25 etc.
Num 11:8
Gen 4:21
Gen 24:22 etc.;
Num 31:50
Gen 37:2
P
P
Num 1:20 etc.
Deut 28:37
Deut 17:18
Gen 41:2 ff.
P
Exod 34:6; Num 14:18
Exod 18:8; Num 20:14
Gen 24:22; Num 7:13
etc.
P
Gen 35:11
Deut 8:4 etc.
Deut 31:27
Exod 34:6; Num 14:18
Gen 41:1 ff.;
Exod 1:22 etc.
Isa 50:3; Job 26:6
2 Chr 4:17; Neh 12:28
1 Sam 16:12 etc.
ie
P
Pp
Jer 8:7; Cant 1:10 etc.
1 Kgs 5:13(4:33);
Ps. 51(50):9
2 Sam 17:28 etc.; Ezek
4:9
Ezek 10:2, 7
P
Judg 19:19
P
Ps 49(48):5 etc.
Ezek 26:13
Ezek 16:11 etc.
Jer 20:10; Ps 31(30):14
Chr 7:20
eh 9:32
a eg te he ee a
"O OOs FG SS
IMPACT OF THE TORAH 19]
onpaola myn Num 10:5 etc.
oKANpOT pd yNAOos * nv nw Exod 33:3 etc.;
Deut 9:6 ff.
2. Lexicon
1 Chr 15:28 etc.; Ezr 3:12
ff.
Prov 29:1
There is no concrete evidence that the translators possessed either
dictionaries or word lists. Thus, when attempting to determine the
meaning of a word, they resorted to various sources of information.
These ranged from exegetical traditions, the context, etymology, post-
biblical Hebrew, the Aramaic language, to the translation of the
Torah. The latter was often consulted when the translators encountered
difficult Hebrew words which also occurred in the Torah,!4 as
exemplified by the following examples (see further the examples
listed by Flashar and Seeligmann [see n. 12]):
1. Deut 32:42 JAN MYID WRW
Judg 5:2A My yID2
Num 5:18 yd)
Judg 5:2B mynd 9193
dm KEemadys apxdvtTuwv €x6puv
ev Tu ApEao8ar dpynyous
Kal dToKaAvyet
dtrexkadug6yn dtoKddAupwa
The A and B texts in Judges are based on different interpretations of
niy75 y193, both of which are reflected in the LXX of the Torah.
2. Lev 5:4 . onow2 x02
ROD
Num 30:7 mnowy XDIN
Ps 105(106):33 wnNw2 XDI
-_
Staote AAovoa Tots yelAEov
... SLAOTELAT
KaTG THY StaaToOANy TY XEL-
hEwy auTis
Kal SLéoTEeLAEV Ev Tots yxel-
AEOLV aUTov
NUI occurs elsewhere only in Prov 12:18 a013 - Aé-yovTeEs.
3. Deut 4:27 7507 "nn
Jer 44:28 oo ty
Gen 34:30 oe
1 Chr 16:19 —
Ps 105(104):12 =. oe
dALyou AprOpai
ddALyooTol
4)
4s
The same phrase is rendered differently elsewhere. !?
4. Exod 1:11 NuDdon(7) *7WV
(ras) wéAELS (Tas) dxupas
12 Thus already Flashar, Septuagintapsalter and Seeligmann, Isaiah, 48.
13 Deut 33:6 eon rn AM «- Kal ZoTw Todds ev dpv6y0s; Deut 26:5, 28:62 vyn *nno -
év dpr6uas Bpayel .
192 CHAPTER TWELVE
2 Chr 8:4, 6; L712 . ue - a“ “
Elsewhere nvD0» is rendered by teplxwpor (Iv MID0n 2 Chr 16:4) and téXrets (2
Chr 32:28).
5. Gen 49:14 ondwnon pa - dvd péoov Tav kKANpwv
Ps 68(67):14 onow Pa - ? -
Elsewhere the word occurs only in Judg 5:16 (B: Sryoptas, A: poodabaty; A in v.
15: yetAéwy).
6. n2DxD in Judg 19:29 is rendered by ppd xatpa in manuscript A in
accordance with Gen 22:6, 10. Elsewhere the word occurs only in Prov
30:14, where it is rendered differently.
7. Gen 49:6 Vou - veupokoTéw
Josh 11:6, 9 ke aM
Elsewhere the Hebrew root occurs only in 2 Sam 8:4 = 1 Chr 18:4.
8. Deut 27:8 mint... nand) - oadids odddpa
2077 ANA (NR
Hab 2:2 mndn by axa. - Kal oadus étl truEtov
The unusual translation of Habakkuk, which makes little sense in the context,
depends on Deuteronomy.
3. Quotations and allusions
Quotations from and allusions to passages in the Torah occurring in the
later books of the Bible were often phrased in the Greek in a manner
identical with the Greek Pentateuch:
1. Num 35:33 PONT DX PIN NT OD... PINT NN 97 N27
Kal ov pt) dovoKTovyonte THY yv ... TO yap alwa
TOUTO dovoKTovet Thy yhv
Ps 106(105):38 = o°073 porxn nm
Kal E€dovoxTovyGn 4 ‘yA Ev Tots ata
govoxtovéw (LSJ: to pollute with murder of blood) does not occur elsewhere in the
LXX. As LSJ does not list other occurrences beside the LXX, the agreement between
the two texts is remarkable.
2. Exod 22:1 xx? NNN ON
éav 5€ €v TH Stopvypate evpe Oj
Jer 2:34 One NINN. x?
otk év SLoptypaotv ebpov altos
Elsewhere &:6pvypa occurs only in Zeph 2:14 (j12n) and Stoptoow reflects 1nn (4
times).
IMPACT OF THE TORAH 193
3. Lev 16:13 (N53 NX) NWP JY 7051
Kal KaAUwBEL FY ATpLS Tod BupLdpaTtos
Ezek 8:11 my MwA 73Y ANI
Kal 7) atuls Tot @uptdwatos dvéBatvev
adtyuls occurs elsewhere only very rarely in the LXX. Tly is rendered mainly by
vede dy.
4. Gen 32:29 ox OV nw
évloxuvoas peta Beot
Hos 12:3(4) OTP NX TW
éviaxuaev pds 8edv
12:4(5) N20 ON TW"
Kal €vloxuoev pEeTA ayyéAou
The Hebrew root does not occur elsewhere in the Bible.
5. Num 25:3 sve bun) Seq tne
Kal €Tedk€aOny Iapand Tu BeeAhe ywp
Num 25:5 mvp 5ya> o-nxin
TOV TETEAEOLEVOV TU BeEekde yup
Ps 106(105):28 nv) 2y29 x7
Kal €TeA€oBnoav TH Beedkdheywpo
Note the unique interpretation of the verb in the LXX (to be consecrated to), for
which cf. also Hos 4:14 (mw7p) and Deut 23:18 wip - Tehkeaddpos .
4. Influence on the exegetical level
The contents of the Greek Pentateuch often influenced the wording of
later translations on an exegetical level.!4 Two examples follow:
1. In Jer 1:6; 4:10; 14:13 and 32(39):17 ‘7 "178 TX has been represented by
5 dv S€oTroTa Kkupte.!° 7nKX (alas) in this verse has been derived from
m7N in Exod 3:14 (a central verse for biblical theology) and rendered
in accordance with the LXX of that verse: TAX WRK WTR - Eye Ele O
wy.
2 Prov 24:28 qy72 on FY AN OK
ut) tobt Wevdys udpTtus ETL aov ToAL THY
The translation of this verse is based on the exegesis of 01m as ‘false,’ mainly on
the basis of the ninth commandment in Greek:
Exod 20:16 (13) 7pw ay qyna myn x»
ov WevdopapTuUptGELs KaTa TOU TrAnOLOV Cou
uaopTtuplav wevdr
14 Soe especially Seeligmann, Isaiah, 45-46.
15 Jn 4:10 only manuscript 26 reads o wv. For a discussion see Tov, Jeremiah and Baruch, 24.
194
Deut 5:20 (18)
CHAPTER TWELVE
NW TY VID myn ND
ov BevsopapTupyGEts Kata Tod TANGloV Gov
Laptuplav Wwevsry
CHAPTER THIRTEEN
THE REPRESENTATION OF THE CAUSATIVE ASPECTS OF THE
HIPHIL IN THE SEPTUAGINT
The causative aspects of the hiphal may be represented in the LXX in
any of the following ways:!
I. Verbs which bear no special features?
II. Causative suffixes
WI. Auxiliary verbs
IV. Reversal of the causative action
Only limited data are available for the relative frequency of each of
the above-mentioned techniques. In some cases the availability to the
translator of certain Greek verbs brought about the use of one technique
rather than another one (e.g., the existence of verbs ending with -dw,
-Uvw, etc.). It is not impossible that for this reason the translator of
Amos 8:5 pw 2773777 7D°X 7>UPN> rendered the second verb with
weyaddvat, but the first one with tottjoat uikpdv ((o)utkpvvetv is used
infrequently in the LXX in a causative sense).
In some translation units a given hiph4l form was consistently
rendered by a single technique, but in others alternative techniques
were used. Thus in some books 12°0(7)_ is rendered by both dya@6wu,
ayadvvw, dyadeov/dyaGda Trotéw, dya0atroléw, as well as by verbs which
! For brief discussions, see H.G.S. Thiersch, De Pentateuchi versione alexandrina libri tres
(Erlangen 1841) 150-153; M.L. Margolis, “Studien im griechischen Alten Testament,” ZAW
27 (1907) 238; Helbing, Kasussyntax, passim; Seeligmann, Isaiah, 55; P. Katz, “Zur Uber-
setzungstechnik der Septuaginta,” WO 2 (1954-59) 269-272; J. Soisalon-Soininen, Die
Infinitive in der Septuaginta (AASF B 132,1; Helsinki 1965) 134-135; J.R. Busto Saiz, La
traducci6n de Simaco en el libro de los Salmos (Textos y Estudios “Cardenal Cisneros” 22;
Madrid 1978). For the Vulgate, see B. Kedar-Kopfstein, “Die Wiedergabe des hebrdaischen
Kausatives in der Vulgata,” ZAW 85 (1973) 196-219.
2 Helbing, Kasussyntax, 80, n. 1: ‘composita with kata- tend towards a causative
meaning’ (e.g., KaTaoknvéw, kaTaoTetSw). Thackeray, Grammar, 289 makes a similar remark
on composita starting with ék- (e.g., éxdoBéw, €Eauaptdvw). Although these statements
may be true with regard to kata - and éx- in late Greek, their relevance to the LXX cannot
be substantiated easily.
196 CHAPTER THIRTEEN
have no external features representing the causative meaning. Note
further dyaSotrotéw and ev Tovéw for 2°v(*)a in the same verse (Num
10:32) and the alternative representations of hiph@l forms in the two
versions of Judges:
Judg 16:26 ~1wnm (Q) A... kal tofnaov ymAadraal pe
B... Kal bnAadyow
Judg 17:13 2-0” A... tyya8otro(noev
B... dya@uvet
It has yet to be determined whether certain translators prefer a given
technique.?
I. Verbs which bear no special features
The technique most frequently used for representing the causative forms
of hiphal is the use of Greek verbs that express a causative idea. Thus
Nea, PI, TI and 72y7 are often rendered by ¢épw and its compound
forms. This group is not discussed in detail.
Il. Causative suffixes
The causative aspect of the verb is often expressed by causative
suffixes, such as are used also in secular Greek: w, -€w, -du), -bw, (Ca,
ACW, -avw, alvw, -Wvw, -EvW.
An examination of the occurrence of these verbs in the LXX is of
interest for the understanding of translation technique, since there often
exists a close relationship between the equivalence of the hiph4l form
and a causative Greek verb on the one hand and the equivalence of a
related Hebrew noun/adjective and its Greek rendering on the other.
For example, the pairs 2°077 - dya06wu (d-ya6tvw) and 210 - dyade are
closely related. In this case, as often elsewhere, the equivalence 1°u77 -
dyaQ6w is probably based on that of 210 with dya6éc, because dya@dw is
not evidenced in the Greek language before the time of the LXX, and
therefore may have been coined by the translators of the LXX (cf. n. 7).
The dependence of the newly coined causative verb on another
equivalence is even more evident in the case of dBatdéw, a hapax
legomenon in the Greek language,° occurring in Jer 49:20 (29:21). The
3 Note the many references to Isaiah, Jeremiah, and Job in group IIL.
4 See A. Debrunner, Griechische Wortbildungslehre (Heidelberg 1917) 89-140; E. Mayser,
Grammatik der griechischen Papyri aus der Ptolemderzeit 1, I (2d ed.; Berlin/Leipzig 1936)
126ff. On the confusion in manuscripts of the LXX between different causative endings,
see Walters, Text, 117-121.
> For details, see Tov, Jeremiah and Baruch, 46-47.
THE CAUSATIVE ASPECTS OF THE HIPH‘IL 17
related dBatoc¢ occurs in the LXX mainly in the second part of Jeremiah
(chapters 29-52 of the LXX), and therefore the use of dBatdéw for own
(hiph4al of onw) in 49:20 (29:21) is based on a rendering which is at home
in that unit.
Presumably equivalents of this type were determined by the
individual translators, while other renderings were borrowed from
other translation units. In fact, the actual procedure was probably
simpler: the equivalence of the whole word-group (root) (y)97 with the
word-group (stem) kaKk- was made the basis of several equivalents used
throughout the LXX.
Non-exhaustive examples are provided below for each of the groups
of causative suffixes together with related equivalences in the parallel
column (e.g., 2°17 - EkKevdw, 72-1 - Kevdc). Only one reference is listed
even if more references are available. Presumed neologisms® are
denoted with a star (*).
1. -dw
dBatéw" own Jer 49:20 aBatog Taw
(2921)
ayabdéuw* 0C)A Jer 32 (39):41 dyaOdc 20
SiKAaLOW 2 T¥n 1 Kgs 8:32 SL KALOG PTX
EKKEVOUW en Cant 1:3 KEVOG 27
€X\aTTOw wynn Num 33:54 EAATTWV pun
(€E)epnpdw ana 2Kgs 19:17 Epnyoc man
GavaTow mnt Gen 38:10 Odvatoc nin
Pautactéw NDOT 2 Chr 26:15 @8aupactdég ND]
KaKOW yan Exod 5:22 KaKOG vg
KaTAKEVOw™ Pid Gen 42:35 KEVOG 27
KAaTAOKNVOW yDwn Ps 7:6 OKTY- yDw
WaTaLéw phatate Jer 23:16 UdTALOG ?an
TATELVOW Dwi 1-Saim 2:7 TATIELVOG Ow
vwdow aha bty Job 39:27 vys- 2}
_ >i bg Isa 28:29 vy- I
2. €UW
Baotrtebw pont Judg 9:16 Baotretc qn
© That is, to the best of our knowledge, the word under consideration was coined by the
translators (or a preceding generation) in order to express words or ideas which, in their
view, could not be expressed adequately by existing Greek words. Our observations are
based on the evidence listed by LSJ, together with LSJ, Supplement. The very notion of
‘neologism’ is subject to the limitations and doubts such as have been described by Tov,
“Compound Words.”*
198
elpnvetw
TOVNpEVOUAL
UTEpToa-
ve voudat
,
3. -€W
dvovew
doe Béw
ddatpéw
€haTToOvoé w*
KaTa-
KANpOVvoLLe w*
KANpovoléw
KANPOVOLE Ww
vopobe TEW
TPWTOTOKE W
UTE pndavew
x
4. -dtw
EKKANOLACW
éEexkAnora cw
TAE OVA CW
5. -bw
év80w
Evioxytw
KATLOXUW
6. -(Cw
aKkoutT((w™
dvabepatlfw
évwtlCopar*
KaToLUKl(w
Koudl Cw
dpklCw
o>wn
yn
tq
yw
yw
on
min
o-yon
Sonn
ehighiy
WNT
iw"
rary
Be 2
TH
Papi
Duden
aaa
weds
nn
>In
yw
oni
PIN
pown
2271
yvrawi
CHAPTER THIRTEEN
1 Kgs 22:45 = el pn
Gen 19:7 TOVNp6G
Neh 9:16 UTE pY}bavoc
Acs
2 Chr 20:35 dvouoc
Job 9:20 doer
Lev 2:9 ddalpeya
Lev 8:29 agalpeua
Exod 30:15 = €da TT wv
Deut 31:7 KAT poc,
KAPOVOLLLG
Prov 13:22 KAT|pOG,
KAnpovoula
Num 14:24 — kdtjpoc
kAnpovouta
Exod 24:12 — vduog
Jer 4:31 TPWTOTOKOG
Neh 9:10 UTE py} pavoc
Lev 8:3 ékKAnola
1 Chr 28:1 éxKAnola
Num 26:54 = hefwv
Gen 3:21 évdupa
Ezek 27:9 loy-
Jer 8:21 lax”
Ps 51(50):10
Num 21:2
Gen 4:23
Gen 3:24
1 Sam 6:5
Gen 24:37
aKovw
avd0eua
wTtlov
ol KOG
KoU@ocG
5pkoc
oy
v7
1
yw7
YP7
aN
mn
pun
Tom
aa
Mn
oe
wT
Prat
ile
7
wind
2m
>In
YAW
on
rags
jDwn
2p
nV IwW
dp0pl Cw
TOoTL Cw
oodl Cw
oTreppaTticw
UTEpaotl Cw
gwttCw
7. AVW
BAXaoTdvw
éxkBAaoTdva
éEauaptdva
8. -alvw
Enpatvw
TapaTLKpalvw
9. -Dvw
ayabtvw
AETITUVO)
Le yadvVvw
LLakpvvw
10. -dw
yevvdw
TLULAW
Additional note: Intransitive verbs used transitively in the LXX
THE CAUSATIVE ASPECTS OF THE HIPHIL
pow
npwin
o-onn
yon
pa
NT, 7,
ae PRT ca
TART
Mp
Ne OT
Ww
wean
ann?
yon
Pagal
yan
JNA
7 n
797
Gen 19:2
Gen 2:6
Ps 19(18):8
Lev 12:2
Zech 9:15
passim
2 Sam 23:5
Isa 55:10
Judg 20:16B
Neh 9:33
Jer
51(28):36
Deut 31:27
Ruth 3:10
2 Sam 22:43
Gen 19:19
Qoh 8:13
Gen 5:3
Lev 27:8
(5p8poc
TOTOG
aoddéc
oTTé 0LLa
dome
DOG
BX\aoTtéc
BrXaoTtdc
auapt-
auapt-
Enpos
TILK POG
ayabdéc
AETITOG
Weyas
Hakp-
yCyYVoULat
Tut
pl ,anw)
pwn
oon
uN
rela)
WN 12]
Nx
nnax
NOM
YW)
199
Since the suffixes of this group were used often in the LXX, as in secular
Greek, to denote causative actions, a further development took place in
the LXX. The translators not only employed existing causative verbs
and coined new ones, but they also used existing intransitive verbs that
ended in -eUw, -Gw, -dvw, etc., as if they were transitive. The internal
logic of the translators’ Systemzwang thus led them to distort the
meaning of existing Greek verbs. E.g., Bactke Uw, usually denoting ‘to be
king,’ ‘to rule,’ in the LXX as well as in secular Greek, is also used in the
LXX as an equivalent of 7°7nn ‘to make king’ (e.g., Judg 9:16, 18 and
7 This rendering, also occurring frequently in Psalms, derives 1707 from 77; cf. Walters,
Text, 150-153.
200 CHAPTER THIRTEEN
passim in 1-2 Kings). Several examples of this type were mentioned by
Helbing, Kasussyntax, 75-80, e.g., €€apaptdvw - x°onn (frequently in 1-2
Kings, e.g., 1 Kgs 15:26), katacknvéw - Dun.
III. Auxiliary verbs
In Hebrew, the causative aspect of the verb is expressed either by the
verb itself or by an auxiliary verb, mainly mwy.8 The Greek translators
similarly expressed the causative aspect of the verb often by an
auxiliary verb, mainly totéw (this use is also evidenced in secular
Greek).
(1) tovéw
(a) Totéw + adj./adv.
(a) The adj./adv. is used elsewhere in the LXX to render the
same Hebrew root: dkovotov trovéw - y°nwi; dkovaTdc - YNwW], etc.; passim,
e.g. Jer 50 (27):2 1-nwr() - dkovoTd Toiyjoate.
BEATLOV TrOLE Ww Bd opey Jer 26(33):13 = BeATiwv 0
€'YKOTIOV TTOLE Ww yn Isa 43:23 € 'YKOTIOG yi
KaAAL Ova TrOLé w Ton Jer 18:11 KQAOG able)
KQAOV TOLE w Pea Isa 1:17 KaA0G 210
LLaKpav TOLEW mn Prov 5:8 LLaKkpdv PIn7
LLLKPOV TOLEW Popa Am 8:5 LLLK POG 1D;
dALyooTOUG ToLéw =n Ezek 29:15 dALYOOTOG pun
dALyouc TOLE Ww bya Jer 10:24 dAlyor pun
TOAUY TIOLEW man Prov 22:16 TTOAUG a7)
TOpPPu) TOLEW ena Job 11:14 Td6ppw nn
TAXY TOLEW piel Isa 8:23 TAXUG plprp.
UWNAOV TOLE Ww lat Prov 17:19 vynAdG 12)
(8) The adj./adv. is not used in the LXX to render the same
Hebrew root:
ATOM TOLEew vw Job 34:12
evrAaBH TOLew PT Lev 15:31
KATAKOTIOV TOLEW = TN? Job 16:7
KEVOV TTOLE Yoni Isa 32:6
TAECOVAGTOV TOLéw = =1277 Deut 30:5
8 Eor biblical Hebrew, see Gesenius-Kautzsch, Grammar, 145, n. 1.
THE CAUSATIVE ASPECTS OF THE HIPH‘IL
(b) trovéw + verb
ddyelv Toréw RIT
TOLEW GUapTadvEeLY NUTT
TOLew ElLae Betty ae
TATHOAL TOLEW tn
TETOLOE VAL TOLEW = Mo?
TOLEW OULLBVaL mph
Totéw UmAadtrjoar = wrnn
(c) Tovéw + noun
APEOLVY TOLEW bnwn
ExBOATV TOLEw on
See further:
Ka8aplLouOv... Yay
TOLEW
KapaTeTaopata miere
TOLE W
OLOONV TOLE Ww Pit
Job 5:18
Isa 29:21
Job 14:3
Isa 42:16
Jer 28(35):15
Jer 32(39):23
Judg 16:26A
Exod 23:11
Jon 1:5
Job 7:21
Am 2:8
Lev 19:27
(d) compound verbs ( -trovéw)
ad yabortroré w* yon
KQKOTIOLEW van
TEKVOTIOLEW hae ay
(2) Other verbs
(a) Verb + noun
dvd TAVOLV min
S(t
Tpoodoklav Yawn
SCSwpLt
Com SS Lt mnn
otrdviov elodyw Vpn
dyadva tapéyw = AN?
ETA YW VETOV Yon
Ka8apav TL6npL =P
Num 10:32
Gen 31:7
Jer 29(36):6
Isa 28:2
Isa 66:9
Isa 57:15
Prov 25:17
Isa 7:13
Gen 7:4
Job 11:13
GNVEW IND
auaptdava NOM
eloépyouat x2
TATE Nat
Tré TOLOa non
oupBalvw ny,
UmrAadaw win
Apeouc meh)?
éEKBa AW 207 (ib.,
v. 12)
cf. dya8dv TroLéw, TOLEW
... dya06v/dyabd (passim)
cf. TOLEW ...
(passim)
AVATIAUOLG
Tpoodoxla
Coon
vETOG
KaKov /KaKd
migehiARgarta)
halla
Sieg
mle)a)
202 CHAPTER THIRTEEN
(b) Verb + prep. + noun
TpOG XapLV ona Prov 7:5 TA TPdG pon (Ezek
éuBddAAW Xd piv 12:24)
el¢ Bac.itéa Onn 2 Chr 36:1 = Baorde te 720
Ka@loTLL
elc Baciikéa ona 1Sam11:15 Bao.devc qn
xplw
(c) Verb + adj./adv.
yakpav até yw pnt Gen 44:4 WaKkpdv Pann
d€tov dtoBalvw awn Job 11:6
8(KaLov 2-73 «Job 27:5; 32:2 Slkaroc "Ty
atodatvw
S{katov kplvw Ptxn = Prov 17:15
TlOnut elvan
doe Br pwn Job 32:3
dava.buc
UdloTapat Tn Prov 21:29
(d) Verb + verb
doeBety &SdoKm wwan Job 10:2 doe Béw ywr
S(Swpt Tppt ~—s- Jer 36(43):20 tredvdAaypéva 7175
dUAGOGELV
TATELYWOAG nw Isa 26:5 TATELVOW nwa
KaTa yu)
dnl tiva Wevdry) IDA Job 24:25 wetSoc ats
AE 'YELV
IV. Reversal of the causative action
The causative action expressed by the hiphal was sometimes reversed
in the translation:?
Isa 39:4 -*NTYINID OPN NR? Ww DT TT RD
Kal olk €oTLy Ev TH OlKW Lou 6 otk EfSooav
Cf. the parallel stich.
Isa 43:26 Fn? nodwi lm DIn
ov &€ pyiodyri kal KprOuipev
Isa 49:26 Onw2 nx FI NX NNT)
Kal dd yovTa. ol OALavtés oe Tac OdpKaG avTiv.
9 CEC. Rabin, “The Ancient Versions and the Indefinite Subject,” Textus 2 (1962) 60-70.
CHAPTER FOURTEEN
DID THE SEPTUAGINT TRANSLATORS ALWAYS UNDERSTAND
THEIR HEBREW TEXT?
The (correct) understanding of the biblical text is an abstract concept.
We do not understand all words in MT, and therefore modern trans-
lations often suggest alternative renderings of individual words, add
question marks, or note that the translation is conjectural (see, e.g., the
notes in NJPST). Furthermore, ancient translators should not be judged
according to our standards, but must be viewed within the framework of
their own world. Turning, then, to the question posed in the title of this
study, we are not focusing on renderings which are mistranslations
according to our standards, but on renderings which show the
translators’ ignorance of words through an analysis of the inner
dynamics of the translation. That lack of knowledge may be reflected
in various types of renderings, especially in conjectural translations.
Conjectural translations must be understood within the framework of
the translation process, in particular with relation to the choice of
equivalents. The whole process of translating in antiquity is often
conjectural, for, to the best of our knowledge, translators had no lexica
or word-lists at their disposal. They therefore had to turn to other
sources of information: the translators’ direct and living knowledge of
Hebrew and Aramaic (including their etymological understanding of
these languages; see TCU, 172-180); exegetical traditions; the context;
and for the later translators, the translation of the Pentateuch (see
Tov, “Pentateuch”*).
The present study focuses on conjectural renderings. The recognition of
such a rendering is not certain, because it is always possible that it
reflects an exegetical tradition or a Hebrew variant. If these caveats
are taken into consideration, several types of conjectural renderings may
be recognized:!
1 According to some scholars, translators sometimes simply omitted difficult words when
they did not know how to render them: Ziegler, Untersuchungen, 52-53; Allen, Chronicles 1,
61-62.
204 CHAPTER FOURTEEN
1. Untranslated words
2. Contextual guesses
3. Contextual manipulation
4. Reliance on parallelism
5. Employment of general words
6. Etymological renderings
1. Untranslated words
One group of renderings demonstrates beyond doubt that at least some
words in the Hebrew Bible were unknown to the translators. These are
words which were left untranslated because the translators did not
know their meaning. Most of these words are objectively difficult,
because they are hapax legomena in the Bible or in the book under
consideration. Probably the translators hoped to return to the
transliterated Hebrew words and to replace them with Greek
translations, or else they expected others to do this (see Tov,
“Transliterations”*). Within the realm of the biblical translations,
these transliterations are found especially in the ‘LXX’ of 2 Kings and
in the sections and fragments attributed to kaige-Th.?
Examples of individual words which were left untranslated because
they were unknown to the translators are listed in Tov, “Trans-
literations,”* and some are repeated here in their respective contexts:
Judg 5:7 ON WeD PWD YIN
LXxA éE€XitTev dpaCwv év Tu *Loparr
Judg5:16 = oonswan pra naw an)
LXx4 tva Tl ot KdOnoal dva Léoov tiv Loodabary
Judg 8:7 O7IPI07 RI (WaTAN PS? NX ODWws7 nx 77)
Kal €v Taic Bapkovuinu
Judg 8:16 OPII7 NI (9277 PSP ART WV PT NX 4779)
Kal Taic Bapaknvi
1 Kgs 5:25(11) wna? 2D (Deon 7D PX OW)
Kal LaxLp Tu olkw avTOt
2Kgs 8:15 2Dnn npn
Kal €AaBev TO Laxya
1 Chr 21:20 oxannn wy 12 nYIAN)
kal Téooapes viol avtov pet abtod pEeOaxafiv
2 The anonymous reviser who produced these two translation units preferred to leave
some difficult words untranslated rather than to indulge in translation guesses.
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 205
As a rule, unknown words were transliterated in their exact Hebrew
form, including prefixes and suffixes, e.g.,
Judg 5:22 ryvax nat naan
LXxA apaSapwd dSuvativ a’Tod
Ezek 41:8 220 279002. m9 cen
Kal TO OpaeX Tov olKOV VisoG KUKAW
Ezra 8:27 37993)
kal Kadoupn xpuoot
1 Chr 28:11) 57201 (1°72: N1...7071)
Kal THY CaKyw avTod (cf. v. 20 LXX)
All these transliterations reflect Hebrew words which are either
hapax legomena (in the Bible or a given book) or were understandably
problematic for the translators.
In Tov, “Transliterations,”* 77 words are listed which were left
untranslated in the LXX (once or more). A further 32 common nouns have
been treated as proper nouns, probably because they were not known to
the translators. A similar list is provided there for kaige-Th. Since the
translators did not know the meaning of these words, it is conceivable
that also other words may have been unknown to one or all of the
translators.
2. Contextual guesses
Since the preceding section demonstrated that several words were left
untranslated, it should not be hard to accept that in other cases the
translators resorted to contextual guesses.
a. Recurring patterns
Some Hebrew words were understandably difficult for the translators,
and if in such cases we meet different renderings in accordance with the
different contexts, it stands to reason that the translators adapted the
translation of the ‘difficult’ word to the different contexts.
A case in point is the translation of 17x.? This word, which occurs
some 30 times in the Bible, is usually translated as ‘palace.’ The word
occurs rarely in postbiblical Hebrew, and this situation may account for
3 For details on the renderings of this word, see R.P. Blake, “Khanmeti Palimpsest
Fragments of the Old Georgian Version of Jeremiah,” HTR 25 (1932) 254 ff.; P.J. Heawood,
““Armon and ‘Aram,” JTS 13 (1911-12) 66-73; Seeligmann, Isaiah, 52; G.B. Caird, “Towards a
Lexicon of the Septuagint, I,” JTS 19 (1968) 460-461.
206 CHAPTER FOURTEEN
the wide range of its renderings in the LXX showing that the trans-
lators were unaware of its meaning, using the context as their guide.
The translation equivalents which come closest to the meaning of
the Hebrew are BaolXetov (‘palace’) in Prov 18:19 and dudodov (literal-
ly: ‘a block of houses surrounded by streets’) in Jer 17:27; 49:27(30:16).
At the same time, we meet the following general equivalents:
vadg (‘temple’)
Jer 30(37):18 = aw HOW DY JIN)
kal 6 vadc Kata TO Kplya alto KabeSet tar
olkog (‘house’)
Isa 32:14 WO) 7
olkot éyKaTane ELL pe VOL
The following renderings probably reflect contextual guesses:
évavttov (‘opposite’)
2 Kgs 15:25 (Q) 7207 M3 [99K PINwWI W197
Kal étatatev avtov év Yapapela Evavttov olkou
ToU BacLtAéwe
TOALG (‘city’)
Isa 34:13 (a°770) ANAK ANY
dvadvuoet ele Tac TddELG aUToV
dvtpov (‘cave’; hapax in the LXX)
1 Kgs 16:18 qPnm nea pox OX RIM
Kal elotopetetat ele Avtpov Tov ofkou Tot
Bao.réwe
Also the following two equivalents referring to specific architec-tural
structures reflect such contextual guesses:
Bdapic (‘tower’) Ps 48(47):4, 14; Lam 2:5, 7; 2 Chr 36:19.
tupyoBaptc (‘fortified tower’?)
Ps 122(121):7. Jr>myanxa aw (42-n2 aw 7°)
Kal evOnvla év tat¢ TupyoBdpealv cov
The representation of 11278 as ‘land’* may reflect contextual exegesis
(cf. especially the parallel pair }17x//y.X in Mic 5:4), but the frequent
occurrence of this rendering may also indicate the existence of a lexical-
exegetical tradition:
yii (‘country’?)
Jer 9:20(21) IPMN XI (127M Nw 7? >>)
eLorjAGev ele Thy yi bysiv
4 Tt is not impossible that the graphic similarity of 7278 and 777% somehow influenced
the present rendering.
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 207
xwpa (‘land,’ ‘country’)
Amos 3:9
Amos 3:10
Amos 3:11
Mic 5:4 (5)
ON PINT MYON 2V7 TIIWRI MIN FV IWlNwA
atayyelvAate xwpaic év ’Acouploie kal étl Tac
xopac THe AlyvTTou
DTN WAIN Tw oN
déuklav kal Tadattwplav év tate ywpate avTisv
TMA Way
Kal StaptayhoovTat al yupal cou
WMI] JT 7D (US7NI N17 7D)
Kal Stay émBry etl Tv xwWpav vv
This exegetical tradition differs from the equally frequently occurring
translation @eyéAta (‘foundations’) in similar contexts in the Minor
Prophets (and elsewhere): Isa 25:2; Jer 6:5; Hos 8:14; Amos 1:4, 7, 10, 12,
14; 2:2, 5.
The mere variety of the renderings, especially within one
translation unit, shows the translator’s uncertainty with regard to the
meaning of the word. Very often the different equivalents are selected
on the basis of their respective contexts:
‘pv (usually taken as ‘hill’) in Jeremiah:
3:21
729
32
T1212
14:6
4:11
yaw avaw 2y Ip
hut ek vyerrAéwv hKovoby
A voice was heard from lips.
(o°°-5v explained from 75w ‘lip’); similarly:
app? ow vy RW
Kal dvddAaBe él yerAéwy 8prvov
ow Dy JPY CRW
dpov ele evdeiav Toc dpbadpotc sou
Lift your eyes to the plain (based on Aramaic,
as in Num 23:3).
oO ttw wa 72IN2 ow 9D by
éttl wdoav dtexBodAnv Ev TH Epryw TA8ocav
TANALTIWPOUVTEG
Upon every pass(?) in the desert destroyers
came.
pow 2 19 OND)
dvot dypto. €otTnoav él vdrac
Wild asses stood on vales.
Dov ny ny
Tvetia mAQVIHCEWS
. a wind of scattering.
208 CHAPTER FOURTEEN
“av is translated as follows in Isaiah (note the parallelism):
41:18 nwpPd ... OP Dw
Opéwy ... TESLWY
49:9 gs 7 ne pe
d60i¢ ... TolBoic
The precious stone onw is identified in different ways:
TIPAOLVOG Gen 2:12
od pS.ov Exod 25:7; 35:9
ovapdy8oc Exod 28:9; 35:27; 39:6(36:13)
BnpvAALov Exod 28:20
COOOL tChr 29:2
mY, or its Qere form mvt (‘horror’), poses no special problems for the
modern lexicographer, yet seems to have been difficult for the
translators:
dvaykn (‘punishment,’ ‘pain’?)
Jer 15:4 YONA M532 259 AYN? Onn
Kal tapaéuow altouc ele advaykasg Tdoate Tatc
Baotrelae THe yc
StaoKkopTLopég (‘scattering’)
Jer 24:9 yux nD? 292 AVA? AVI? ann
Kal Wow attoug el¢ Staokoptiopov ele Tdoac
Tac Baothelac THe yc
Staotopd (‘scattering’)
Deut 28:25 PrN m2 799 AyW nM
Kal €oy év Staotopd év tdcatg tate Baotdelatc
TNG Yh
Jer 34(41):17) pax ma2nn 599 ay? oDnKX cnnn
Kal wow vpde etc S:aotopav doatc tatc
Baotrelate m6 yc
ékoTaotc (‘astonishment’)
2 Chr 29:8 mprw 7) aw? Ay on
Kal €5wkev avtouc elc Exataowv Kal el¢
ddaviowov kal ele cuptoudv
édtil¢ tovnpd (‘bad expectation’)
Isa 28:19 Ww Pad AW PA WM
.. €OTAL EATILE TOVHpd: UdbeTE AKoveLV
Also the following conjectural renderings of an-w/nmw (‘pit’) are based
on their respective contexts:
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 209
Ps 119(118):85 mmw ott) nD
Godless men dug pits for me.
Sijyrjoavtdé Lo. tapdvopor ddoreoxlac
Transgressors told me idle talk.
Jer 18:20 "W512 AMwW AD
They dug a pit for my life.
ouvedddnoav pryata KATA THE WuxT|e Lov
They spoke words against my soul.
Jer 18:22 (Q: aAMw) AMmw MD
They dug a pit.
évexelpnoav Adyov
They formed a plan.
In these verses, the meaning of nm-w/nmw was not recognized and the
word was taken as am (‘conversation’). This rendering obviously
changed the meaning of the context in which the verb did not fit any
more. 17) had little to do with ‘conversation,’ and accordingly the
translators adapted the translation of the verb to their respective
objects:° &inyyoavTo (‘they told’), cuvedddnoav (‘they spoke’), évexel-
opnoav (‘they undertook’). Probably the relative frequency of the occur-
rence of the words influenced the translators, since 7n°v together with
mt occurs much more frequently in the Bible than an-w. In Ps 57(56):7,
however, the translator recognized nmw which was easily recognizable
in the context. The conjectural nature of the renderings in Jeremiah is
underlined by the fact that the same phrase was rendered differently
in two adjacent verses (18:20, 22).
b. Isolated instances
The almost identical verses Isa 18:2 and 18:7, which contain several
difficult words and forms, have been rendered in different ways,
reflecting different attempts of solving lexical problems.
MT v. 2 v.2 v.7 MT v. 7 (when
different)
ON TPOG EK
3 €8voG haov ov
qwnn LLETE WPOV TEOALLLLEVOU
ebay kal Eévov Kal TETLALEVOU
245 must have been known to the translators as can be established at least in the case
of Ps 7:16; 57(56):7; 94(93):13.
210 CHAPTER FOURTEEN
28
ov Aaov Kal até Aaot oy
XVI Kal yadheTIOv Le ydAou
NY 7 tice atTov amd Tob viv
ARI éTEKELVA Kal elc¢ Tov alava ypdvov
“1 €@voc €@voc
?P Wp avé ATILOTOV é€ATIL Cov
7012) Kal KaTaTeé- kal KaTaTeTIaTTWLEéVOV
TaATNWEVOV
NID WR vov 6 €oTiv €v pepe
on ol Totauol TOTALLOU
WW THG YG THE xWpace avtot
In v. 2, ]wind is rendered contextually by petéwpov (‘haughty’), while in
v. 7 etymologically by te@\tppévov (‘squeezed’); likewise, in v. 2, 0119 is
rendered contextually by Eévoc (‘strange’), but in v. 7 etymologically by
TeTLALEevou (‘peeled’). x17 79 is taken as an interrogative pronoun in v. 2
(probably read as 8177) or W712 as in Aramaic), but as a7 Tov viv in Vv.
7. Likewise, in v. 2, 47N27) is taken in a local sense as é€téKeLva
(‘beyond’), but in v. 7 chronologically as kal el¢ tov aldiva xpévov (‘and
to the eternity’). 1p 1? is ‘hopeless’ in v. 2 (a4vé\tLoTov), but ‘hopeful’ in
v. 7 (EAT Cov). INI2 WR is vov in v. 2, but 6 €otiy év péper (‘which is in
the part’?) in v. 7 (this rendering is probably based on a separation of
INI2 in two words, -2 and 1x1, or the like, even though the nature of the
second element is not clear).
The differences between the two translations probably reflect the
translators’ hesitations rather than an attempt to distinguish artifi-
cially between two or three different peoples, for such a differentiation
is not borne out by the evidence.®
3. Contextual mantpulation
In some cases the avoidance of a difficult word is subtle, and therefore
more difficult to recognize. We submit that the translators sometimes
knowingly manipulated the Hebrew consonants in order to create words
which would fit the context better than the words of their Vorlage,
Ov. 2 probably refers to two peoples described as E6voc peTéwpov kal Eévov Aadv kal
xadeTtév and another one described as €6voc dvéAtLotov Kal kataTeTaTHpLEVoV, both
depicted in negative terms. Likewise, v. 7 probably refers to three peoples, of which the
first one is described negatively (Te8\ppévov kal TeTLALEVOV), the second one positively
(eydAov ), and the third one in mixed terms (éATlCov kal kaTatretTampévov).
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 211
either because the Vorlage was not understandable to them or because
the translator made certain adaptations in the wake of other changes
or mistranslations. Such renderings do not reflect real variants, but
rather ‘pseudo-variants,’ that is, Hebrew readings which existed only
in the translator’s mind and not on parchment (see TCU, 162-171). The
alleged manipulations are based on the translators’ paleographical
understanding, for it must have been known to them that certain
Hebrew letters were graphically so similar that they were often
interchanged in Hebrew sources. Therefore a translator who could make
no sense of a word when written, let us say, with a daleth, would have
been strongly tempted to render it as if it were written with a resh. The
assumption of such paleographical manoeuvring is objectively
conditioned by the occurrence of lexical or other difficulties. Examples
have been discussed in TCU, 162-171. One such example is repeated
here, and a few are added.
Jer 31(38):8 MT VAN? NIP) AT AO] WW | PAX wD ANI?
AL DN ae ani
And I shall gather them from the farthest
parts of the earth, among them the blind
and the lame, the pregnant woman, and the
one in labor, together, a great multitude
shall return hither.
LXX Kal ouvdEw abtove dt €oxdtov THC yije év
EopTH dacek Kal TeKvotrottjon SxAoV TONY
Kal dtootpéwouaty ade
And I shall gather them from the farthest
part of the earth at the feast of Pesach,
and you will give birth to a great
multitude, and they shall return hither
(implying: N05 7123).
The Greek translator had a text in mind that differed completely from MT,
ascribing the return of the Jews from the exile to the time of Passover (cf. T to Cant
1:1 referring to Isa 30:29). The great difference in meaning between MT and the
LXX is based on a relatively small difference in consonants and vowels. Once the
words ‘among them the blind and the lame’ (MT) had been read as ‘at the feast of
Pesach,’ the context was completely changed and the translator was impelled, as it
were, to conceive of several details in the verse in a way different from MT. In
particular, the words ‘the pregnant woman and the one in labor, together’ (777
yin? n32°}) did not suit the new context. This caused the translator to introduce a
second verb, parallel to the first one, by vocalizing 712°) instead of n7?>).
212 CHAPTER FOURTEEN
Furthermore, he represented neither 777 nor 13n°. The upshot of this maneuvering
was a rendering kal texvotrotyon SxAov ToAvv (and you will give birth to a great
multitude). The translator’s Vorlage of the whole phrase was, as it were, 121”")
217) 277 II) MOP Ivina yox "NIV any? 7. The existence of that reading and
its vocalization must be strongly doubted.
In some cases the translators felt at liberty to manipulate the
consonantal text, disregarding prefixes and suffixes:
2 Chr 35:13 MT nw7?8I7 O77) nwo. wa
They boiled in pots, in cauldrons and in
pans.
LXX Himoav év tole yadketore Kal €v Tol
AEBNoLY " Kal EvodwOn
They boiled in the copper vessels and in the
pots, and it succeeded.
2% (pan) of MT is a hapax legomenon, while the related nn2x occurs three times
in the Bible and mn?x once. The word was probably unknown to the translator,
who derived it from the verb nx (‘to succeed’), disregarding both the internal
division of the verse and the prefix and suffix of the word (cf. Allen, Chronicles, I,
p. 61). The translation, which does not suit the context, was based on a cluster of
consonants in which the translator recognized the meaning ‘to succeed’ without
entering into details regarding the precise form of the word.
In the following examples, the translator read his Vorlage wrongly
in such a way that he introduced, as it were, linguistically incorrect
forms. We submit that these forms, too, were found only in the mind of
the translator and not in his written text.
Jer 2:20 MT mt AVS AX
You bend like a harlot.
LXX éxet dtayvonoopar év Ti topvela ou
There I shall be spread abroad in my
fornication.
= m1 *TY RMN
Staxv@roopat must probably be retroverted as *nyxnX (presumably éxet was
added contextually). But one notes that the retroverted *7v8nx creates a
morphologically unlikely form (790¥X) whose meaning is unclear. In spite of the
unlikely form, only this reconstruction seems to account for the unusual
translation.
See further Jer 6:25 analyzed in detail in TCU, 76-77.
In the following example, the translator read the consonants
wrongly:
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 213
Gen 47:31. Tenn wRt Oy 2XIW? INnNw)
And Israel bowed upon the head of the bed.
Kal mpooeKUunoeV Iopana etl TO dkpov T1¢ pdBSou
auTou
And Israel bowed upon the top of his staff.
From the context it is clear that in MT a bed (mittah) is meant rather than a staff
(matteh).’” In fact, when the word occurs next in the story, the translator identified
it as ‘bed’ (48:2 kXLvn). In 47:31, however, he fails to identify the word because the
text had not mentioned explicitly that Jacob was ill or in bed. Furthermore, matteh
occurred twice in chapter 38, so that the translator’s error is understandable.
Neither the translator nor a subsequent reviser corrected the error.
Prov 8:1 N22 [AN WAN) NIPN ANI NIA
Does not wisdom call, does not understanding raise her
voice?
LU THY oodlav knpv&erc tva dpdvnole cou UTaKovoT
You will announce wisdom in order that understanding
will obey you.
The translator wrongly took X77n as a second person masculine verb rather than a
third person feminine governed by non. This understanding introduced an
unwarranted ov into the translation which changed the whole context.
4. Reliance on parallelism
Reliance on parallelism is a form of contextual translation, treated
here separately. As a rule, reliance on parallelism is a stable means of
determining the meaning of words, but the decision whether or not to
turn to parallelism remains subjective and the recognition of different
types of parallelism requires different renderings. Some of the
equivalents mentioned above reflect such a reliance: e.g. Mic 5:4
(pow / /y~ x), Isa 41:18 (mypa//orow), Isa 49:9 (a--»w/ /o7307). When the
translator presumably relied on a parallel word, producing an
acceptable rendering, we cannot be certain that this was the case, as
the choice of equivalents may also have derived from his knowledge of
the Hebrew language unrelated to the context. Only when invoking the
parallel word created unusual equivalents (or different equivalents in
several contexts), can such reliance be identified with confidence.
Reliance on parallelism may take two different forms:
7 For a detailed analysis, see J. Barr, “Vocalization and the Analysis of Hebrew among the
Ancient Translators,” (VTSup 16; Leiden 1967) 1-11, esp. 3.
214 CHAPTER FOURTEEN
a. Choice of parallel Greek word
The choice of equivalents on the basis of the parallel word has been
recognized especially in the translation of Isaiah (see Ziegler,
Untersuchungen, 9 ff).
Isa 5:11 nwin // 7pr2
TO Tpwl ... TO OWE
Isa 21:4 pen Aw // 17>
tT) Kapd&la pov ... 1) ux Lov
Isa 59:10 nwin // oy2
é€v weonuppla ... év pecovukTlw
Although the rarely occurring w1 was known to some translators, the translator
of Isaiah did not know its meaning. He used three completely different renderings
in accordance with their respective parallels. Possibly 21:4 is irrelevant if the
translation was based on a different Hebrew reading w1 (metathesis).
In the following renderings, pi3y1 is resolved according to the
parallel word, in 7:19 according to o-v70 and in 55:13 according to 1570.
Isa 7:19 oxryyin bom // oy2on vpn
Kal €v Taic Tpwydatc Tuv TeTpav Kal el¢c Td OTMAaLa
Isa 55:13 O77 729° JNIOA NNN wd Avy? pyyyin nnn
Kal dvtl TH¢ oToLBc dvaBroeTat kuTdpLacoc, dvTl 6é
THE KovucTG dvaBroeTat pupolyn
b. Repetition of the parallel word
More secure than the aforementioned technique was the repetition of
the parallel word when translation of a given word was difficult, for
example when the word was a hapax legomenon or rare (see Ziegler,
Untersuchungen, 20).
Jer 8:16 ... 2p (7010) NIN (yaw! 770)
PWV ... PWUAG
mn does not occur elsewhere, but the related 1m1 occurs also once in the Bible.
Jer 10:20 Mey OPM CAR FWY AVI PR...12MI) -AaN 731 (TIw 77K)
Kal mdoat al S€ppe.g Lov ... ToTI0G Tuv SE ppewv Lov
mn occurs elsewhere 8 times.
Isa 2:16 (A7on7) Nrow (7D 9y7 wewan) nYaX (9D 2)
TAotov ... TAolwv
nvDwv occurs only here.
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 215
5. Employment of general words
Ignorance of a word is often disguised by the use of general words which
the translator considered to be somehow fitting in the context (e.g. ‘to
do,’ ‘give,’ ‘arrange,’ ‘prepare’). It is not easy to prove that a given
rendering reflects such a contextual guess, but that assumption is likely
when the Hebrew word is objectively difficult. For example:
Ps 84(83):7 a7.) Noy? nND73: Oh
The early rain will also cover (it) with blessings.
Kal yap evAoylac Sacer 6 vopobe Tuy
For the lawgiver will also give blessings.
Elsewhere the translator of the Psalms knows the meaning of the verb Avy (in Ps
71(70):13; 109(108):19, 29, for example, where the context makes it clear that the
covering of a dress is meant, he uses teptBddAdovat or the like). In this verse,
however, ‘he got himself thoroughly lost,’® for he ‘missed’ a few words in the
immediate context, and in the section quoted above he wrongly took 471» as
‘lawgiver.’ Accordingly an etymologically correct rendering of the verb may have
been considered inappropriate by the translator. In any event, he contented
himself with a general equivalent (dace - ‘he will give’).
Of special interest in this regard is the use of Tapaoxevd {w (‘to
prepare’) as a general equivalent in Jeremiah.’ In the first two of the
following examples, the translator must have known the Hebrew verbs,
but he probably could not locate appropriate renderings; in the next two
examples, the Hebrew verbs probably were unknown to him.
Jer 6:4 manda m>y wp
Sanctify war upon her.
Tapaokeudoaobe ET abTiy elc¢ 14d€LOV
Jer 46(26):9 29709 Yan
Rage, O chariots.
TapaoKevdoate (kal kataokevdoate LXX4) Ta &pyata
Jer 12:5 O-olom MX WAMNn pt
How will you complete with horses?
TUG TapaoKkeudon Ed imTOLG
Jer 51(28):11 o-sna nan!
8 Thus Barr, Comparative Philology, 249.
9 This verb occurs five times in Jeremiah and six times elsewhere in the LXX.
10 Elsewhere the verb occurs only in Jer 22:15—also its translation there (tapofbvn )
should probably be regarded as a translation guess.
A reconstructed Vorlage 1127 is not impossible, but methodologically difficult. Ont only
is 127 graphically remote from 1725, but also the resemblance to the other three cases makes
the likelihood of a contextual guess greater.
216 CHAPTER FOURTEEN
Sharpen the arrows.
TapackevddeTte Ta TOEEUWATA
Beyond Jeremiah tapaokevd¢{w is also used as a general equivalent: 1 Sam 24:4;
Prov 15:18; 24:27(42); 29:5.
2Chr 14:49 ond (nx. nan nr)
Kal Ta €l6wda
2 Chr 34:4. gynnm (o->van mina)
Kal Ta bumAd
2 Chr 34:7.) myonn 2D)... (o°9097) OWWNA ... nINIIn7)
Kal TavtTa Ta Lisnrd
The meaning of 0°39n (probably ‘sun pillars’ used in idolatrous worship) was
probably conjectured from the respective contexts. Elsewhere the word occurs five
times.
The translator of Psalms used tapdcow (‘to cause disorder’) for a
whole range of Hebrew verbs, the meaning of some of which may have
been unknown to him: 71n0, 373, 072, ton 29n, wyr, 10, 2ow, now, nox, OV,
wwy (see Barr, Comparative Philology, 252).
6. Etymological renderings
a. Root-linked renderings
Many translators rendered all occurrences of a given Hebrew word,
element (e.g. preposition), root or construction as much as possible by
the same Greek equivalent ( stereotyping). It is probably true to say
that from the outset a tendency towards stereotyping was the rule
rather than the exception. The system of stereotyping was an integral
part of the translation technique and it originated from the approach
that the words of the Bible should be rendered consistently in order to
remain as faithful as possible to the source language. This type of
translation created a consistent representation of whole Hebrew word-
groups (roots) with Greek words also belonging to one word-group.
While this root-linked system had its origin in a certain conception of
translation technique, it was also used in connection with difficult
words. If such a difficult word has a recognizable Hebrew root, it was
sometimes rendered by a Greek word belonging to a Greek stem that
elsewhere rendered other Hebrew words belonging to the same word
group (root). The Greek word does not necessarily carry the same
meaning as the Hebrew word, but other words close to that Greek word
are used elsewhere as renderings of Hebrew words close to the Hebrew
DID THE SEPTUAGINT TRANSLATORS UNDERSTAND THEIR HEBREW TEXT? 217
word under review. In our view the following examples show that
translators sometimes resorted to root-linked renderings when the exact
meaning of the Hebrew word was not known to them.
nixwn, ‘kneading trough,’ occurs three times in the Bible. In Deut
28:5, 17 it was translated by éyxatdAetppa, and in Exod 12:34 by dvpaya.
*EykatdAetypa (‘remnant’) conveys no meaning which comes close to
‘kneading through’!? and it was merely chosen because the root of the
Hebrew noun, 1xw, was translated elsewhere by (éy)katadel(trw.
oip> (‘all that exists,’ ‘substance’) was translated in Gen 7:4 by
éEavdotaoig (AM...: dvdotepa) and in Gen 7:23 by dvdatnya. These two
words have to be taken as ‘rising,’ ‘ressurection’ and not as ‘living
being’!3 and both are based on the frequent equation orp - (€€) aviotnpL.
In most instances, however, it is very hard to know whether an
etymological rendering reflects a concern for the consistent represen-
tation of Hebrew word groups with equivalent Greek word groups or
whether it disguises the translator’s ignorance of the exact meaning of
the word. For example, both Mv! and atrottvayya occur only in Judg 16:9.
This rendering is obviously based on the translation of 7’ in v. 20 with
dtottvdoow. The same verb renders 193 in Lam 2:7.
b. Etymological guesses
Reliance on etymology is a known procedure for tranlators, and such
reliance is called conjectural when the translation is based on a certain
manipulation of the consonants, sometimes involving disregard of
prefixes or suffixes. In all cases the Hebrew words involved are
understandably difficult. Several examples have been analyzed in
detail in TCU, 172-180.
Translators were often ignorant of the meaning of the words in their
Vorlage and this ignorance led to several conjectural renderings.!4 In a
12 Pace LSJ, s.v. which quotes no other source for this meaning than the LXX of
Deuteronomy.
13 pace H.S. Gehman, “Adventures in Septuagint Lexicography,” Textus 5 (1966) 129.
14 cf. Allen, Chronicles, 59: ‘It is not difficult to perceive that now and then the
translator came across words whose meaning he did not know and could not discover. He
seems to have had three distinct methods of dealing with the situation ... The first and
most common expedient is guessword.’ Gerleman, Job, 19: “Cruces interpretum are often
evaded by the Greek translator by dividing the text in his own way ... he often commits
mistakes in regard to the significance of individual words and phrases’; Seeligmann, Isaiah,
56 ’... we shall only give a modest selection of those passages where an interpretation based
on misunderstanding let the translator to make free explanatory additions.’ Cf. also A.
Bludau, Die alexandrinische Ubersetzung des Buches Daniel und ihr Verhdltnis zum MT (BSac 2,
2-3; Freiburg 1897) 87-96 (‘Falsch tibersetzte Stellen’).
218 CHAPTER FOURTEEN
world without lexica, this situation should not cause much surprise.
Only very rarely the translators were sophisticated enough to leave
words untranslated (group 1 above).
The amount of conjectural translation in the LXX is probably
relatively extensive, but the real number can never be determined.
CHAPTER FIFTEEN
COMPUTER-ASSISTED STUDY OF THE CRITERIA FOR ASSESSING
THE LITERALNESS OF TRANSLATION UNITS
IN THE SEPTUAGINT
WRITTEN JOINTLY WITH B.G. WRIGHT
Generations of scholars have commented upon the contents of the LXX,
both on individual words and passages and on the translation character
of the version as a whole. Such remarks referred to various aspects of
the translation, its exegesis, language, inner-translational variants,
translation technique, and its text-critical value for the Hebrew text. In
all these aspects it is recognized that analyses and evaluations of
individual words and passages are essential, but that final judgment is
often not possible without reference to the same or similar phenomena
or equivalents in the translation unit under investigation.
The study of translation technique and that of text-critical
evaluation are two distinct areas, but in practice they are closely
connected, as becomes evident from an evaluation of the text-critical
procedure. Those who are interested in the text-critical evaluation of
the LXX attempt to find out whether the deviations of the LXX from
MT reflect Hebrew variants. Very often scholars merely decide ad hoc
whether or not a certain deviation of the LXX reflects a Hebrew
variant. However, such decisions can be made properly only on the
basis of a thorough knowledge of the character of the translation unit
as a whole. One would like to know whether the translation is
considered faithful to the Hebrew or free, or somewhere between the
two. If one has a preconceived view that the translation is literal, upon
encountering a substantial deviation, one’s first thought is that the
deviation resulted from a different Hebrew Vorlage.
Likewise, if the translation is considered paraphrastic or free, one’s
first thought would be that the deviation resulted from the translator’s
free approach to his source text. In both cases there is exaggeration,
because free translations also reflect variant readings (e.g., Proverbs),
220 CHAPTER FIFTEEN
and similarly literal translators also allow themselves some free
renderings (e.g., the translation of Ruth, ascribed to kaige-Th). Never-
theless, generalizations regarding the character of the translation are
helpful.
In the past scholars allowed themselves to make generalizing
statements on the translation character of the individual books and
these statements have been used in the text-critical evaluation of the
translation units. Such statements are found in monographs dealing
with individual translation units, and on the basis of such analyses one
also encounters conclusions on the translation character of all the books
of the LXX.!
The characterizations of the translation character of the LXX are
often impressionistic and unsatisfactory, since they are based on the
scholar’s intuitive understanding of the translation character. Thus, if
a scholar felt that lexical consistency rather than inconsistency
characterized a certain translation unit, he provided several examples
of the former while adding a few examples of the latter as ‘exceptions.’
Another scholar could make just the opposite statement by providing a
list of examples of the latter. Almost all descriptions are partial, and
hardly any absolute numbers are given. Hence, in most of the extant
analyses of the translation techniques employed in the books of the
LXX, it is not the examples that count, but the scholar’s impressions.
There has been an awareness that the terms ‘literal’ and ‘free’ are
imprecise, especially in Barr’s discussion.2 In TCU, Tov outlined five
criteria for describing the degree of literalness of a rendering (and
translation unit), four of which can be expressed in statistical terms.?
The very use of statistics is meant to provide an element of precision in
this area in which so far scholars relied too much on mere impressions.
To date, few attempts have been made to express the literalness of a
translation unit statistically,4 but mention should be made of R.A.
Martin’s list of criteria for determining ‘translation Greek,’ worked out
in detail for the translation units in the LXX, as well as for the New
1 E.g., Thackeray, Grammar, 13 ff.; R.A. Kraft, IDBS (Nashville, 1976) s.v. “Septuagint,”
pp. 813-814; Tov, TCU, 63.
2). Barr, The Typology of Literalism in Ancient Biblical Translations (NAWG, I. Phil.-Hist. KI.
1979, 11; Géttingen 1979) 279-325.
TCU, 54-66: 1. consistency, 2. the representation of the constituent elements of
Hebrew words by individual Greek equivalents, 3. word-order, 4. quantitative represen-
tation, 5. linguistic adequacy of lexical choices.
ce Marquis, The Translation Technique Reflected in LXX-Ezekiel, unpubl. M.A. thesis,
Hebrew University Jerusalem 1982). For an earlier attempt, see Y.-J. Min, The Minuses and
Pluses of the LXX Translation of Jeremiah as Compared with the Masoretic Text: Their
Classification and Possible Origins, unpubl. diss. Hebrew University (Jerusalem 1977).
CRITERIA FOR ASSESSING LITERALNESS 221
Testament and secular Greek.° Some of these criteria might also be used
for measuring the literalness of a translation.© Furthermore, the studies
of N. Turner, I. Soisalon-Soininen, and R. Sollamo of several words and
word groups measure the relative literalness of the translation units in
which they occur.’
This study does not measure literalness in general, but is lumited to
examining the consistency of the translation. Literalness is an abstract
designation, and consistency is one of its main exponents. Since
consistency can be measured, this article suggests it as the key to
measuring literalness.
An attempt is made here to measure the translation character of the
individual translation units with fixed sets of criteria. This is a
computer-assisted sample study in which a few such criteria are used
involving books of the LXX which represent different degrees of
literalness.
The database used is CATSS,® containing, among other things, a
running text of the LXX and MT, element by element, verse by verse, etc.
It records ‘formal’ equivalents of the LXX and MT (as if the LXX were
translated from MT) in col. a, and the ‘presumed’ equivalents of the
LXX and its presumed (retroverted) Hebrew Vorlage in col. b.? This
SRA. Martin, Statistical Evidence of Semitic Sources in Greek Documents (SCS 3; 1974).
© See the list of criteria in the first chapter, pp. 5-43. Although some of Martin’s criteria
are helpful also for the present investigation, his samples from the LXX itself are too scarce
since his main interest was not inner-LXX investigation like the present study, but the
relation between original and translated Greek. See B.G. Wright, “A Note on the Statistical
Analysis of Septuagintal Syntax,” JBL 104 (1985) 111-114.
N. Turner, A Grammar of New Testament Greek, UI, Syntax (Edinburgh 1963) 332
(relation between kal and 8€ in the LXX); I. Soisalon-Soininen, Die Infinitive in der
Septuaginta (AASF B 132, 1; Helsinki 1965), especially 176 ff.; Sollamo, Semiprepositions.
Mention should also be made of other studies in which statistics are provided for one book
only (sometimes the Pentateuch, especially in the studies of Soisalon-Soininen). These
statistics are helpful, but they are of limited value in the comparison of the various
translation units: R.A. Martin, The Syntax of the Greek of Jeremiah, Part I: The Noun, Pronouns,
and Prepositions in Their Case Constructions, unpubl. diss., Princeton Theological Seminary
1957; I. Soisalon-Soininen, “Der Gebrauch des Genetivus Absolutus in der Septuaginta,”
Proceedings of the Fifth World Congress of Jewish Studies, TV (Jerusalem 1973) 131-136; id., “Die
Wiedergabe einiger hebrdischer, mit der Praposition be ausgedruckter Zeitangaben in der
Septuaginta,” ASTI 11 (1978) 138-146; id., “Der Gebrauch des Verbes exein in der
Septuaginta,” VT 28 (1978) 92-99; id., “Die Konstruktion des Verbs bei einem Neutrum
Plural in griechischen Pentateuch,” VT 29 (1979) 189-199; id., “Die Wiedergabe des be
Instrumenti im griechischen Pentateuch,” in: Glaube und Gerechtigheit. Rafael Gyllenberg in
memoriam (Vammala 1983) 31-46; J.H. Sailhamer, The Translational Technique of the Greek
Septuagint for the Hebrew Verbs and Participles in Psalms 341 (Studies in Biblical Greek 2;
New York 1991).
8 See Tov, “Computerized Database.”*
? For the definition of these types of equivalences, see Tov, TCU, 142 ff.
Zon CHAPTER FIFTEEN
Vorlage is retroverted merely for selected words, and in most cases not
for the words covered by this study. In the course of recording the
equivalents of MT and the LXX, various types of notes have been
incorporated into the database which are of importance for the study of
translation technique, some of which are described below. The
following description refers to and exemplifies criteria for the study of
the literalness of translation units, together with the statistical data.
The text included in the edition of Rahlfs provides the textual basis
for the study, while textual variations included in the apparatus are
disregarded. Accordingly a margin of error must be taken into
consideration, which, based on the comparison with the volumes of the
Gottingen Septuagint, does not exceed 1%. The absolute numbers for the
occurrences of the equivalents listed below have been culled from the
CATSS database with the aid of the Oxford Concordance Program
(OCP).!9 This program enables the search of all the occurrences of a
particular target word in any translation unit. In this study, the
translation unit usually comprises an entire book except for Psalms and
Job.
The following translation units are included in the samples
examined: Numbers, Deuteronomy, Judges A, Ruth, 1 Samuel, 2 Samuel,
2 Kings, Ezra, Nehemiah, Job 1-5, Psalms 30-65, Proverbs, Qohelet,
Canticles, Jeremiah, Lamentations, Ezekiel, the Minor Prophets
(individually and as a group), and Sirach. For example, the OCP
program located all the occurrences of the preposition -1 in Canticles
(68), of which 61 are rendered by év, 2 by ws, 1 by the genitive case, 1 by
the accusative case, 1 by the dative case, 1 occurrence is €&w rendering
rina and one occurrence has no Greek equivalent. Thus, -12 is rendered by
év in 90% of the cases.!! This figure can then be utilized in comparisons
of the rendering of -2 by év in other translation units. The use of
percentages also helps to minimize the differences in the size of the
books. For this same reason, the statistics for the Minor Prophets (MP)
are given for each individual book and also for all of them together.
10s, Hockey and I. Marriot, Oxford Concordance Program, Version 1.0, Users’ Manual
(Oxford 1982).
1 For each criterion, those instances where there is no equivalent at all in the Greek
have been subtracted from the total number of occurrences in Hebrew in order to make
the percentages as representative as possible for all translation units. This applies especially
to books such as Jeremiah which had a Vorlage shorter than MT and Sirach for which only
a partial Hebrew text is preserved. This type of computation introduces a certain margin of
error into these percentages, but given the small number of cases where there is no
equivalent in the other translation units, that margin of error is not great enough to change
the final results.
CRITERIA FOR ASSESSING LITERALNESS 225
Five specific renderings have been selected in our examination of the
amount of consistency and hence of literalness/freedom of the
translation units:
1. -1 represented with év;
2. “Drepresented with Ott or 6671;
3. the Hebrew third person singular masculine suffix represented
with attés and éautés;
4. the frequency of prepositions added in the LXX in accordance
with the rules of the Greek language or translation habits;
5. the relative frequency of the post-position particles 8é, pév,
obv, and Te in relation to kat.
The description of each criterion is accompanied by both the
absolute numbers and a table of percentages arranged in order from most
literal to least literal.
1.2-é€v
Probably one of the best criteria for investigating the literalness of
translation units is the rendering of -2 by év creating numerous
Hebraisms (for some data, see the studies of Soisalon-Soininen
mentioned in n. 7). The large number of occurrences in each translation
unit and the wide range of possible renderings make this an important
criterion for literalness.!2
Though ێv is usually the main rendering of -2 in most translation
units, a large number of other possibilities exist:
CHART 1: Renderings of -3
comp. no
=3 by em mpd ta Ta kaTd__cas. ty d and & _ repr.
Num 730 339 47 = 28 3 14 4 57 145 12 9 14 11
Deut 763 404 44 56 o 1 15 4 13: - 133 22 3 11 11
Jes A 557 398 16 6 2 5 5 5 -75 15 1 1 il 6
Ruth 45 31 2 1 8 1
1Sam 712 412 81 27 2 ig 1 8 72 e) ] 1 50
2Sam 459 328 20 8 1 1 54 3 1 1 6 3
2 Kgs 572 465 8 8 I 4 4 36 2 12
Ezr 155 134 1 12 1 1
Neh 207 161 4 1 3 22 3 1 S) 8
12 It should be admitted that most instances of -2 reflect the instrumental use of the
preposition, which is rendered well by év, and the statistics are particularly meaningful for
occurrences of -2 which are not instrumental. It may be assumed that these instances are
equally distributed in the books of the Bible, but if it be determined that they are not, a
factor of slight imprecision should be taken into consideration.
224 CHAPTER FIFTEEN
Job 56 815 1 S) 15 1
Ps 299 204 12 34 2 2 22 1 1
Prov 400 116 22 20 2 3 6 15 131 3 2
Qoh 146 135 1 1 1 2
Cant 68 61 2 2
Jer 997 560 66 42 3 3 Z 4 5 109 5
Lam 104. 78 3 4 1 1 1 10
Ezek 1126 676 54 66 3 2°15: 16 8 137 3. 14
Hos 116 76 7 8 3 1 1 15
Joel 40 23 4 2 2 6
Am 112 64 10 9 1 1 3 4 14 1
Ob 2p. -12 4 i 5 1
Jon 23 8 2 1 1 1 5
Nah 29 =«:18 1 3 1 6
Hab 38 8624 3 2 8
Zeph 51 32 2 2 3 1 i: <2 1
Hag 36 —s «16 2 11
Zech 180 128 2 9 1 1 2 2 26 Zz
Mal 5A 43 3 5
MP 704 444 36 42 4 4 6 8 8 108 6
Sir 553: : 327 9 8 1 1 13 7 95
Miscellaneous renderings
évavt -/évwt- 2, éket 2, adverb 5, rapa 4, tpwl 6, rept 1,é@ws 1, Sonar 1, dvd 1,
verb 2, ovv 1, inf. 1, ae 2, bte 4, dtav 6, odk 1
Ste 3, éti 1, \vlka 1, €vavt-/évwt- 13, obk 1, FEw 1, dvr 1, €dv 1,a priv. 1, &¢ 1,
i776 3, €ws 2, towl 2, of 1, Tepl 1
of dv 2, adv. 1, dua 1, hwlka 1, rept 1, obv 1, Tapa 1, verb 1, évavt -/évwtr- 11
Kal obv édv 1, rpwl 1
ob dv 1, molv 2, bép 1, oul 9, verb 3, Evavt-/évwtr- 9, adv. 2, pH 1, inf. 2
ét. 2, €vavt -/tvwt- 5, Ste 1, verb 2, btws 1, dvtl 1, tpwl 4, Evexev 5, Etdvw 1,
trepl 1
Tepl 2, d8e 1,00 édv 1, €vw- 7, mapd 1, woul 4, €Ew 1, ékeT 1, mrplv 2, Ews 1
bros 1, TéTE 2, inf. 1, €Ew 1
verb 1
no repr. 3, verb 3, napa 1, ind 1, Urrép 1
omdte 4, dvev 1, dtav 1, évortr- 3, ZEw 1, verb 1
no repr. 1, dtav 2, €vwi- 2, €Ew 1, verb 14, dvd pécov 1, wept 1, bte 1, adv. 9, bd
2,tdv 2, hwika 9, rapa 5, tolv 1
Kabdtt 2, kabuis 1, doa dv 1
ao eo
CON
1
oir
CRITERIA FOR ASSESSING LITERALNESS Peas)
éEw 1, évl 1
€Ewbev 2, éket 6, étt 1, adv. 3, mowl 1, mapa 1, ovv 1,{va uh 1, verb 2, évavr-
/évut- 5, tepl 8, dte 5, br 2, b5 1
no repr. 1
inf. 4, €vexev 1, ét. 1, bd 2, odk 1, étrdvw 1, &tav 6, éxet 2, €EwOev 1, ob 1, Exet 1,
bs 1, €owdev 1, Hvika 1, ovv 1, dvtl 1, brép 2, adv. 2, tpwl 10, Ste 3, verb 2, tepl
1, €vavt- 1
ouv 1, umd 1, verb 1
avrt 1,el 1
verb 1, évexev 1, avtt 1
Zws 1
éxtevus 1, €vexev 3
évuott -/évavt- 1
bre 1, verb 1, bép 1, €vavt-/évuwt- 1, vvép 1
évavt -/eévwtir- 2
verb 8, €vexev 2,014 2, adv. 6, repl 1, dv 8, ply 3, rapa 1, évavt -/évuntr- 2
The percentages in chart 2 represent the relationship of the total
number of occurrences of -1 to the number of occurrences rendered by év.
CHART 2: Percentage of renderings of -1- év
1. Qoh 92.4% 11. Ruth 70.4% 21. Joel 57.5%
2: Cant 90% 12. «2s 68.68% 22. Am 57.1%
3. Ezr 87% 13. Hos 66.6% 23. Deut 53.7%
4. 2 Kgs 83% 14. Jer 65.5% 24. Hag 50%
5. Mal 79.6% 15. Sir 64.7% 25. Ob 48%
6. Neh 77.7% 16. Ezek 64% 26. Num 47.1%
7 Lam 76.4% 17. Hab 63.1% 27... On 34.7%
8. Judg A 72.2% 18. MP 63% 28. Prov 30.3%
9. 2 Sam 71.9% 19. 1Sam 62.2% 29. Job 27.7%
10. Zech 71.9% 20. Nah 62%
2. °2- dtt and S671
By far the most frequent rendering of °D in its various meanings is 8tt, in
some boo
ks in 100% of its occurrences. The table records all renderings of
°> so that the different proportions can easily be detected. ydp is not a
major equivalent of *D except in the Wisdom books of Proverbs (67%),
Sirach (56%) and Job (38.8%). This may be due to the fact that ydp is a
post-posi
ST. to the
tion word and more literal translators would have preferred
disturbed the Hebrew word order implied by ydp.
226 CHAPTER FIFTEEN
CHART 3: Renderings of °
25 St. SidTe yap Stav édv ox adr" t G\Ad Kal no equiv
Num 124 77 19 2 10 2 3 1 4
Deut 268 140 1 36 4 66 1 9 8 ) 7
Judg A 110 94 a 1 2 3 2 Pe
Ruth 27 21 1 2 2
1 Sam 243 8206 2 1 il 2 8 2 12
2 Sam 162 139 3 2 2
2 Kgs 108 98 3 1
Ezr 24 24
Neh 41 37
Job 21 5 7 1 1 1 3
Ps 115 ~=:104 5 7 1 d
Prov 99 10 59 1 oF 2 2 11
Qoh 87 86
Cant s) i)
Jer 425 302 25 Lod 1 1 13 7 56
Lam 29 21 2 1 1(?) 3
Ezek 202 108 68 1: 13 1 2 11
Hos 69 oe) 26 1 2 1
Joel 32 25 7
Am 26 10 12 1 1
Ob 3 3
Jon 13 10 3
Nah 6 2 4
Hab 16 1
Zeph 17 6 10 1
Hag 3 eS)
Zech 48 9 33 2 2 2
Mal 26 7 17 2
MP 259 109 125 1 2-3 5 3 3 1
Sir 142 38 65 2 1 2 q 26
Miscellaneous renderings
Num el 1, rAdv 1, ob 1, b¢ 1, obv 1
Deut yy 1, obx. 1,4 1
Judg A étrel 1, &:d4 1, bte 1
Ruth ews av |
1 Sam Ste 1,001, 8d 1,el 1,ev td ...1
2 Sam Ste 2, ob 2, obxt 1, el 1, o¥ uh 1
Neh el wr 2, Et. &€ 1
CRITERIA FOR ASSESSING LITERALNESS
Job {va Tl 1, vdv 2
Prov tva 2, obSe 1, Tus 1, 8 1, phtote 1, rds 1
Qoh el py 1
Jer émevdy) 1, wy 1, ot 1, 64 1, 1600 1,4 1
Lam Kalye 1
Ezek el uy 1, kaTd 1,65 4
Hos 8a TodTtTo 2, tls 1, el 1
Am évexa 1, éav pyl
Hab 6é 1
Sir aotep 1, €l 1, oltus 2, rod 1
227
The separation into three columns shows that in some books (i.e.
Nahum, Joel, Malachi), °> was rendered by either 5tt or &t6TL, and not
by any other rendering. All the numbers in the following table represent
the relationship of the total number of occurrences of *2 to the number of
occurrences rendered by dt (col. 1), 8671 (col. 2) or both words together
CHART 4: Percentage of renderings *D - 671 /6.6Tt
(col..3);
9 /6Tt
i: Ezr 100%
2: Cant 100%
3. Qoh 98.8%
4, 2 Kgs 95.1%
5. Neh 92.5%
6. Ps 90.4%
ig 1 Sam 89.1%
8. 2 Sam 87.4%
9. Judg A 87.03%
10 Ruth 84%
11 Jer 81.8%
12 Lam 80.7%
13 Joel 78.7%
14 Jon 76.9%
15 Num 76.9%
16 Ezek 56.5%
17 Deut 53.8%
18 Hos 48.5%
19 Hab 43.7%
20 MP 42%
21 Am 38.4%
»D/8tdTt
Hag 100%
Ob 100%
Mal 70.8%
Zech 68.7%
Nah 66.6%
Zeph 58.8%
MP 48.2%
Am 46%
Hab 43%
Hos 38%
Ezek 35.6%
Jon 23%
Joe! 21.8%
Jer 6.7%
Ruth 4%
2 Sam 1.8%
Judg A 92%
1 Sam 8%
Deut 3%
Ezr -
3/OTL-SLOTL
Mal 100%
Jon 100%
Joel 100%
Nah 100%
Ezr 100%
Hag 100%
Ob 100%
Cant 100%
Qoh 98.8%
2 Kgs 95.1%
Zeph 94%
Neh 92.5%
Ezek 92.1%
Ps 90.4%
MP 90.3%
1 Sam 89.9%
2 Sam 89.2%
Jer 88.5%
Ruth 88%
Judg A 87.95%
Hab 87.4%
228 CHAPTER FIFTEEN
Pipe Zeph 35.2% 2 Kgs - Zech 87.4%
23. Nah 33.3% Lam - Hos 86.7%
24. Sir 32.7% Neh - Am 84.4%
25. Mal 29.1% Num - Lam 80.7%
26. Job 27.77% Prov - Num 64.16%
27. Zech 18.7% Ps - Deut 54.1%
28. Prov 11.36% Ooh - Sir 32.7%
29. Ob - Sir - Job 27.7%
30. Hag - Cant - Prov 11.36%
3. + and }- - autés or Eautés
The most consistent rendering of the third person singular masculine
suffix is either avtés (autod, etc.) or €autds. This equivalence is so
consistent that it is frequently redundant in Greek (such as in nouns
joined by kal). Those translation units which might be termed ‘free’ for
the most part do not render the ending at all.
CHART 5: All renderings of + and 1-
1 Ps abtés €auTos other rel. pron. norepr. no equiv.
ers. pn.
Num 552 416 2 4 2 a7 12
Deut 535 422 6 80 12
Judg A 430 404 4 18 2
Ruth 40 oF 1 1
1 Sam 634. 508 6 43 64
2 Sam 523 475 4 3 4 28 o
2 Kgs 613 57 3 i) 23 19
Ezr 63 6] 1 1
Neh 120 95 i 10 13
Job 54 34 1 3 11
Ps 150 133 2 5 1 7 2
Prov 349 138 50 1 il 131 PP:
Qoh 110 108 2
Cant 60 53 2 1 4
Jer 532 406 4 6 3 39 73
Lam 43 39 1 1 2
Ezek 516 386 3 10 4 84 29
Hos 86 al 2 2 11
CRITERIA FOR ASSESSING LITERALNESS
Am 52 46
Ob 14 13
Jon 25 23
Nah 20 20
Hab 53 4] 3
Zeph A
Hag 10
Zech 92 87
Mal 34 29 1
MP 427 378 6
Sir 604 350 1
Miscellaneous renderings
Num demonstr. 11, €tepos 1
Deut demonsttr. 6
Judg A demonstr. 1, proper name 1
1 Sam demonstr. 4
2 Kgs proper name 2
Neh demonstr. 1
Job YSto¢g 1, éxetvas 1
Prov éxetvos 1, (80s 1, ofSev 1, demonstr. 1
Jer demonstr. 1
Sir demonstr. 6, t8L0s 2, Totodtas 1
229
The numbers in the following chart indicate the relationship of the
total number of occurrences of the endings 1- and 1- to the number of
occurrences of these endings rendered by attos or €autdés. The table
shows a relatively narrow range of variation, since only 6 of the 29
books examined fall under 80%.
CHART 6: Renderings of + and 1- with avtos/éauT6s
Joel
Nah
SOF Oe NS a es Oe
=)
ay
oS
9)
a
=
100%
100%
98.3%
98.1%
97.5%
96%
95.6%
95.3%
95.1%
94.8%
11.
12.
13.
14.
15.
16.
lz,
18.
19;
20:
2 Sam
Ob
Cant
Ps
1 Sam
Amos
Hag
Jer
Neh
Mal
93.1%
92.8%
91.6%
O152%
90.5%
90.3%
90%
89.3%
88.7%
85.2%
84.8%
82.4%
81.8%
79.8%
78.84%
78.7%
68.6%
67.5%
57.49%
230 CHAPTER FIFTEEN
4. Prepositions added in the LXX in accordance with the rules of the
Greek language or translation habits
The LXX adds many prepositions to the Hebrew text, often due to the
requirements of the Greek language. In other instances, the translator
chose to render a Hebrew word or phrase by a prepositional phrase in
Greek, especially in renderings of the Hebrew construct state. A
preposition added in the LXX reflecting these situations is denoted in
the CATSS database by “{..p.”
1. Prepositions added in accordance with the requirements of the
Greek language
Gen 12:15 npn/) Kal eloryayov
Tw /A alTyy
few els
ma {..p els} Tov olkov
ny7p Papaw
2. Prepositions added as a translational technique
Deut 32:1 IN byjwaTa
{...] ék
ay A {..p €k} oTépaTdés
LLOU
Literalness or freedom in a translation cannot be judged when a certain
rendering was almost forced on the translator (such as the example from
Genesis 12). Nowhere is this more evident than in 2 Kings which is
usually estimated to be very high on the scale of literalness, but due to
the nature of the Hebrew of this book is low (‘relatively free’) in this
category. On the other hand, Qohelet contains no renderings of this
type, which suggests that other, probably more comfortable, alter-
natives were at the disposal of the translators. The number of cases
where the translator is more or less constrained to add a preposition
appear to be fewer than those added according to translation technique.
As a result, this category can be used as a criterion for literalness, but it
must be used cautiously in conjunction with the other criteria.
Seemingly anomalous data, such as in 2 Kings, must be analyzed more
closely for the use of prepositions added by the translator in order to
conform to Greek usage. The following chart shows a wide range of
prepositional usage in this category.
CRITERIA FOR ASSESSING LITERALNESS 231
CHART 7: Prepositions added in the LXX
words
& els emt xatd pds amd Sd fuse os & Tapa Urép vd Tepi peta counted
Num LZ 18° 7 8 2 3 29 9 1 8 25629
Deut 1 4 4 1 3 1 1 23033
Judg A 5 29 4 1 I 1 1 6 1 16346
Ruth 1 3 2 2115
1 Sam 4 54 4 3 1 1 2 21406
2 Sam 8 45 2 il 2 1 18226
2 Kgs 26 45 5 1 2 2 19334
Ezr 1 6 t ] 5693
Neh 3 5 dL 8081
Job 1 1962
Ps 2 3 i 1 1 13419
Prov 2 3 1 1 2 1 1 12538
Qoh no occurrences
Cant 1 1 2038
Jer 28 64 8 1 6 1 1 1 pi 1 44122
Lam 2 3 2449
Ezek 22 10 7 8 3 7 1 1 2. 2139
Hos 1 8 1 1 1 2 2 3985
Joel 4 J 1592
Am 1 y 2 1 3262
Ob 2 i 482
Jon 4 2 1 1114
Nah if 1 955
Hab 1 1 L132
Zeph 1 if 1247
Hag 1 2 a, 997
Zech 2 3 1 1 5049
Mal no occurrences
MP By “1 7 Zz 1 yi 2 3 6 2 2 21674
Sir Lo “7 +3 1 6 1 1 1 21654
The percentages listed in chart 8 represent the relationship of the
occurrences of Greek prepositions added in the LXX to the total number
of words in the text.18
13 The total number of words counted is larger than the actual number of words in the
text because OCP counts a]l words which are repeated in the database (such as els and éx
in the examples given above) as different words. Since this limitation pertains to all the
translation units examined, the final outcome is not affected.
232 CHAPTER FIFTEEN
CHART 8: Percentage of prepositions added in the LXX
1. Qoh - 11. Sir 17% 21. Joel 31%
2. Mal - 12. Hab 17% 22. 2Sam .32%
3. Job 05% 13. Ezek 18% 23. 1Sam = .32%
4. Ps 05% 14. Lam 2% 24. Num 39%
5. Deut 063% 15. Nah 2% 25. Hos 4%
6. Cant 09% 16. Jer 25% 26. 2Kgs 41%
7. Neh 11% 17. Ruth 28% 27. Ob 6%
8. Zech 13% 18. MP 28% 28. Jon 62%
9, —Ezr 15% 19. JudgA .29% 29. Hag 8%
10. Zeph 16% 20. Am 3% 30. Prov 95%
5. The relative frequency of the post-position particles 5é, uév, ovv, and
Te in relation to kal
Kal is frequently employed in the LXX as the coordinating word
rendering -1.!4 The preference of Hebrew for coordination with -1
naturally leads to an abundance of kal in the LXX. As a matter of course
then, the more consistent a translation, the fewer coordinating particles
other than xal should appear. The most common and frequently
occurring alternative is the post-position word 6e. In the cases of pév,
ovv, and te only a few translation units contain any instances at all and
these words are most helpful as indicators of the free nature of these
units.
CHART 9: Frequency of post-position particles in the LXX
Kal B€ obv év TE words
counted
Am 312 5 3252
Deut 2013 99 4 1 6 23450
Ezek 3232 or al 32108
Ezr 600 3 2 4 5689
Hab 99 5 1126
Hag 95 ps 2 996
Hos 361 20 3983
Jer 2760 25 32143
14 See especially Turner (n. 7 above); Martin (n. 5); A. Aejmelaeus, Parataxis in the
Septuagint, A Study of the Renderings of the Hebrew Coordinate Clauses in the Greek Pentateuch
(AASF, Diss. Hum. Litt. 31; Helsinki 1982).
CRITERIA FOR ASSESSING LITERALNESS
491
1679
112
2146
19
5S
13
2
3
233
1590
abt
16293
19280
2447
1859
952
8079
25587
482
8529
4555
2108
21385
18216
2038
5041
1246
12233
20219
The percentages in the table below represent the relationship of the
number of occurrences of each post-position particle to the total number
of occurrences of kal in each translation unit.
CHART 10: Frequency of post-position particles in relation to Kal
Kal /&é kal /ovv Kal /TE Kal /év
Qoh - Sir - Zeph - Am -
Nah : Zeph - Zech : Ezek -
Cant - Zech - Cant - Ezr -
Neh 1% Cant - 2 Sam - Hab -
1 Sam 38% Ruth - 1 Sam - Hos -
2 Kgs 39% Qoh - Ruth - Jer -
Ezr 5% Ps - Qoh - Joel -
2 Sam 89% Ob - Ob - Jon -
Jer 9% Neh Neh - Judg A -
Judg A 95% Nah - Nah - 2 Kgs -
96%
234 CHAPTER FIFTEEN
Ezek 1.14% Lam - Lam - Mal -
Zech 1.4% Jon - Judg A - Nah -
Lam 1.6% Joel - Jon - Neh -
Am 1.6% Jer - Joel - Ob -
Joel 1.7% Hos - Jer - Qoh -
Mal 2% Hag - Hos - Ruth -
Hag 2.1% Hab - Hag - Cant 7
MP 2.5% Ezek - Hab - Zeph -
Num 2.9% Am - Am - Job -
Jon 3% MP - Ezek 03% 1 Sam 03%
Sir 3.2% 2Kgs 04% 2Kgs 04% Deut 04%
Deut 4.9% 2 Sam 04% MP 09% 2 Sam 04%
Hab 5% Judg A 1% Num 12% Num 12%
Hos 5.5% 1 Sam 11% Sir 23% MP 14%
Ob 8.3% Deut 19% Deut 29% Zech 18%
Ruth 9.6% Num 2% Ps 38% Ps 19%
Ps 4.7% Ezr 33% Ezr 66% Sir 23%
Job 60.7% Prov 81% Prov 2.6% Prov 61%
Prov 121.% Job 2.6% Job 2.6% Hag 2.1%
6. Conclusion
On the basis of the information in the charts the books can be divided
into the categories of ‘literal,’ ‘relatively literal,’ ‘free,’ and
‘relatively free.’ Some books, however, are difficult to characterize
because of the range of variation from one criterion to another, at least
at this initial stage of the investigation. A good example are the Minor
Prophets (taken both as individual books and as one translation unit).
In the case of the rendering of "> and the third person singular masculine
suffix, the translation of the Minor Prophets is consistent and represents
a relatively literal translation technique, but a relative amount of
freedom is shown in the treatment of -2 and the addition of
prepositions. A relatively large number of post-position particles
appear in the Minor Prophets as well. The translation units which fall
in this category, termed ‘inconsistent’ or ‘indecisive,’ are Sirach,}5
19 Sirach is fairly consistent in rendering -2 and in not adding prepositions, but it is low
on the scale of renderings of *3 and the third person singular masculine suffix, as well as
the use of post-position particles.
CRITERIA TOR ASSESSING LITER ALNESS 235
Psalms,!& Lamentations,!”7 1 Samuel,!® Ezekiel,!9 and the Minor
Prophets. This evidence leads to the conclusion that in some books, the
translator had relatively fixed ways of translating certain Hebrew
words or phrases, while other words were translated with greater
flexibility depending on the context.
The remainder of the books show a relative consistency of
translation (whether literal or free) which can be described on the
basis of the categories listed above. Chart 11 gives the position (first,
second, etc.) of each book in relation to the others for each criterion.
CHART 11: Relative literalness of translation units
3 / STL »5/ “s/f -a/év LAR {.p Kal/8 kal/otv kal/te
Si6TL OTL autaés
StOTL éauTds
Num 14 19 19 26 26 18 ne 6 5
Deut 16 18 20 23 22 3 20 S) 7
Jdg A 8 16 13 8 i 14 8 3 1
Ruth 9 14 12 10 9 13 24 al 1
1 Sam 6 17 9 19 14 17 3 4 5]
2 Sam 7 Ais, 10 9 10 17 6 2 l
2 Kgs 3 19* 3 4 8 20 4 2 3
Ezr 1 19* 1 3 2 vs 5 7 9
Neh 4 19* 5 6 18 5 2 a 1
Job 25 19* ae 29 27 2 26 9 10
Ps 5 19* 7 11 13 2 25 Need 8
Prov 27. 19* 23 28 29 24 27: 8 10
Qoh 2 19* 2 1 3 1 1 1 i
Cant 1 1)" 1 2 I2 4 ] i 1
Jer 10 13 1 Is 17 12 % 1 1
Lam 11 oF 18 7 4 5B T2 1 1
Ezek 15 10 6 15 24 10 10 1 2
16 Psalms is relatively literal in criteria J-[V, but criterion V shows the use of a large
number of post-position particles.
17 Lamentations is relatively free in criterion I and very literal in criterion II while it
fluctuates in the other categories.
More accurate results for Samuel and Kings can be obtained by dividing the books
according to the translation units as defined by Barthélemy, Devancters (1963) and sub-
sequent studies.
Books such as Ezekiel, where the translation character seems to change from one unit
to the next, should probably be subdivided and investigated within the limits of these
changes. Studies such as the present one may be of help in evaluating these changes in
translation character.
Kal /
LEV
Coo Se FP SP PB SP eS eH
—
i)
= | pe |
236 CHAPTER FIFTEEN
Hos 17 9 16 12 21 19 22 ik 1
Joel 12 12 1 21 1 16 13 1 1
Am 20 7 17 22 15 15 12 1 1
Ob 28" 1 1 25 11 21 23 1 1
Jon 13 11 1 2f 5 22 18 1 1
Nah 22 4 1 20 1 11 at 1 1
Hab 18 8 14 16 25 9 21 1 1
Zeph 21 5 4 18 23 8 9 1 1
Hag 28" 1 1 24 16 23 15 1 1
Zech 26 3 1S 9 6 6 11 1 1
Mal 24 2 1 5 20 1 14 1 it
MP 19 6 8 17 19 13 16 1 4
Sir 23 19 pal 14 28 9 19 1 6
is no occurrences
** 100%
A number of books consistently appear high in the tables given above
and are called ‘literal.’*? These are: Qohelet, Canticles, 2 Kings,?1
Ezra, and Nehemiah.** Those books which consistently fall in the top
half of the table, but do not show the same consistency of literal
translation as the above-mentioned books are termed ‘relatively
literal.’ These are: Judges A, Ruth, Jeremiah, 2 Samuel.22
At the other end of the scale, Job and Proverbs appear at the bottom
of the tables in almost every case, and are therefore considered ‘free.’24
Deuteronomy and Numbers, though not demonstrating the same freedom
as Job and Proverbs, consistently fall in the lower sections of the tables
and are thus called ‘relatively free.’
As more criteria are investigated, the group of translations which
we have called ‘inconsistent’ and ‘indecisive’ should become smaller,
and clearer distinctions between ‘literal’ and ‘free’ translators will be
determined.
20 For the specific percentages of each book with respect to the other translation units
see the tables given above.
71 See above for a discussion of the addition of prepositions in this book.
22 Nehemiah falls fairly low positionally on the table for criterion I. However, the
range of variation, as noted above, is so narrow that this position can be misleading. It is
just as important to note that the translator of Nehemiah rendered the third person
ending in Hebrew by attés or €autds in 88.7% of the instances.
23 9 Samuel falls lower in criterion IV for the same reasons as 2 Kings.
4 The one exception is Job in the category of added prepositions, where it stands very
high.
CRITERIA FOR ASSESSING LITERALNESS 257
The groups delimited above agree to a great extent with those
established by Soisalon-Soininen in his analysis of a single item, viz.,
the renderings of the Hebrew infinitive and with Sollamo in relation to
the renderings of Hebrew semiprepositions (see n. 7). Soisalon-
Soininen’s three groups represent a rather general division into free and
literal with Job, Proverbs, Deuteronomy, and Numbers included in the
group which tends toward free renderings, and the books included in the
categories ‘relatively literal’ and ‘literal’ of the present study falling
into his third or most literal group.*? Likewise, the four groups in
Sollamo’s study roughly correspond to the four groups set out above.
Proverbs and Job fall into the group called by Sollamo ‘freely rendered.’
The second category, corresponding to our ‘relatively free’ category,
contains Numbers and Deuteronomy. Sollamo’s third group, our
‘relatively literal,’ is comprised of the Minor Prophets, 1 Samuel, and
Psalms. Her fourth group, called ‘most slavishly rendered,’ includes II
Esdras, Jeremiah, Ezekiel, Judges A, 2 Kings, 2 Samuel, Lamentations,
and Qohelet. The major differences between the results of the present
study and those obtained by Sollamo are the inclusion of Jeremiah,
Ezekiel, and 2 Samuel in the group which we have called ‘literal’
(Sollamo’s ‘slavish’ group).*° 2 Samuel, as can be seen from the tables
given above, is a borderline case and could possibly be considered a part
of the ‘literal’ group. Jeremiah, on the other hand, clearly belongs in
our relatively literal’ group while Ezekiel (as well as Psalms and 1
Samuel in Sollamo’s third group) have too much variation between
criteria to be placed in any one group and have therefore been called by
us ‘inconsistent.’ In an experimental and incomplete study such as this
the basic agreement between our groups and those of Sollamo needs to be
emphasized, and not the differences.
Future studies should expand the list of criteria to include
transpositions, the choice of lexical equivalents, as well as other
ones.2” On the basis of this new evidence, the degree of literalness of
the individual books can then be expressed in statistical terms, to the
benefit of future work on the text-critical use of the LXX which no
longer has to rely on general impressions.
25 Soisalon-Soininen, Die Infinitive in der Septuaginta, 177-178, 186, 189 (see n. 7 above).
26 Sollamo, Semiprepositions, 284-286.
27 For a different view, see the introductory remarks to A. Aejmelaeus, On the Trail of the
Septuagint Translators—Collected Essays (Kampen 1993) 1-2.
CHAPTER SIXTEEN
THE NATURE AND STUDY OF THE TRANSLATION TECHNIQUE
OF THE SEPTUAGINT
The symposium for which this study has been written is devoted to
translation technique. In this century, and especially in the last three
decades, several thorough studies have been written on the translation
technique of the LXX, even though the research of the past century had
required such studies already at that time. Important aspects of that
area have now been analyzed, but at the same time we are still far
removed from a full understanding of the translators’ techniques.
Even if translation technique has not been studied thoroughly in the
past, it has always been of interest. Of special concern were the
techniques used by the first translators, since they had no earlier
models to consult, they had to devise their own. The case of the LXX is
especially interesting, since that translation transposed a Semitic text
into a language which had a completely different structure. Hence, the
LXX translators had to cope with difficult problems, as they had to
locate equivalents for grammatical categories of the Hebrew language
which had no exact or even approximate equivalents in Greek, and
sometimes none at all. For example, the Greek language has no
equivalent for the Hebrew infinitive absolute construction (*n?0? 20>),
or for constructions with °n7 or 401%), or for the combination of wx with
the so-called redundant pronoun ( e.g., 1°?Y ... 1x). By the same token,
Greek contains constructions which do not exist in Hebrew. Thus there is
no equivalent in Hebrew of the genetive absolute construction, and the
Greek verbal system is much more developed than that of Hebrew.
Furthermore, when translating the Hebrew, the translators were often
faced with distinctions required by Greek which were not made in
Hebrew. Thus the translators often had to make a decision between the
modes of the Greek verb, or between its various tenses, such as those of
the past.
What exactly is meant by the study of translation technique has not
been a matter of dispute among scholars, as little attention has been
240 CHAPTER SIXTEEN
devoted to the definition and demarcation of this area. Two
publications illustrate this lack of clarity. The collection of articles
published by S. Jellicoe as Studies in the Septuagint: Origins,
Recensions, and Interpretations (New York 1974) contains a section
named “Text, Translation Technique.’ Several studies in that section are
indeed devoted to matters of text, and under the heading ‘translation
technique’ the editor included a study dealing with anthropomor-
phisms as well as a study of the bisection of books for the purpose of
translating. However, many articles could have been found for the
rubric of ‘translation technique.’ Likewise, the Classified Biblio-
graphy contains a section (paragraph 16) devoted to ‘translation
technique,’ listing mainly articles on composite authorship.
What is translation technique? That term has become a terminus
technicus denoting the special techniques used by translators when
transferring the message of the source language into the target
language. This includes the choice of equivalents, the amount of
adherence to the Hebrew text, the equivalence of Greek and Hebrew
grammatical categories, and etymological exegesis. It also refers to
some of the conditions under which the translation was written and
about which information is included in the translation itself:
cooperation between translators and use of earlier translations. In this
definition revisional activity is not included, although that, too, could
be included under the heading of translation technique.
When reviewing the literature on translation technique, we note
that no relevant section is found in the various publications of Nida, but
Nida is really more interested in analyzing modern Bible translations
than in describing ancient translations. Likewise, Swete, Introduction
contains no section nor even a paragraph on translation technique, nor
does the Introduction by Fernandez Marcos.? Jellicoe, SMS, 314-318,
intended as an update of Swete’s Introduction, does contain a section on
‘translation technique.’ That section, however, merely speaks about the
categories ‘free’ and ‘literal’ in the translators’ approaches, and not
about other aspects of translation technique.
On the other hand, the popular A Handbook to the Septuagint by
R.R. Ottley (London 1920) does contain a section dealing with problems
1 £A. Nida and Ch.R. Taber, The Theory and Practice of Translation (Leiden 1974). E.A.
Nida, “Principles of Translation as Exemplified by Bible Translating,” in: R.A. Brower, ed.,
On Translation (New York 1966) 11-31, esp. 22 ff. where Nida speaks of ‘grammatical
categories.’
N. Fernandez Marcos, Introduccién a las versiones griégas de la Biblia (Textos y Estudios
‘Cardenal Cisneros’ 23; Madrid 1979).
TRANSLATION TECHNIQUE 241
of translation technique, even though that term is not used. In chapter
V (‘The character of the translation: the Greek and the Hebrew’),
Ottley deals with the difficulties in rendering categories of Hebrew
grammar into Greek. His examples refer to the Hebrew and Greek
tenses, relative clauses, the infinitive absolute, the repetition of
elements in Hebrew, and various other peculiarities of both languages.
The discussion is short, but sets out some of the basic problems.
Several thorough studies of translation techniques have been
written in this century, while a first beginning was made in the last
century. At this juncture the difference between grammatical studies
and studies of translation technique should be pointed out. Gramma-
tical studies take the language of the LXX as their point of departure,
treating that language within the framework of the development of
the Greek language as a whole. It is known that the language of the
LXX has been influenced much by its Hebrew source, but it is natural to
treat the LXX merely as a document of the koine language, because the
LXX is such a Greek document. Studies of translation technique, on the
other hand, focus on the techniques used in the translation of the
Hebrew into Greek and when doing so they also contribute to our
understanding of the Greek language. Grammatical studies thus center
on the language of the LXX, while studies on translation technique also
analyze how this special language came into being. A major difference
between the study of language and of translation technique is that the
latter takes the categories of the Hebrew as its point of departure,
while the study of grammar necessarily starts with the categories of
the Greek language. Thus, an analysis of the various renderings of the
causative aspects of the hiph4l, such as carried out in my own study
“Hiphal"* exemplifies the study of translation technique.
On the other hand, scholars interested in the Greek language
compare the forms used in the LXX with the overall picture of the
Greek language in the koine period, and also with the development of
the Greek language over the centuries. In the case of the causative
endings, certain causative verbs used in the LXX are not known from
other sources. This may be a matter of coincidence, since only a fraction
of the evidence relating to the Greek language is known, but it is not
impossible that the LXX translators coined new forms. See Tov,
“Hiphal’* for details.
The study of translation technique started in modern times,to the
best of my knowledge, in 1841. In that year two important studies were
published within a close geographic proximity. In Erlangen, Thiersch
published his De Pentateuchi versione alexandrina libri tres, and
242 CHAPTER SIXTEEN
Frankel issued in Leipzig his Vorstudien. The third book of Thiersch’s
libri tres, named Hebraismi, deals with various idiosyncrasies of the
language of the LXX created by the adherence of the translators to the
Hebrew. This is a first discussion of translation technique in the true
sense of the word, since in every paragraph it also treats the
background of the phenomena in the Hebrew. There had been treatises
on the language of the LXX also before 1841, but these did not
sufficiently take into consideration the Hebrew background of that
language. In this third book, Thiersch discussed the use of pronouns,
cases, prepositions, renderings of the lamed, tenses and conjugations of
the verb, of the infinitive absolute, etc. The remarks in this work,
however, are haphazard, and not based upon much evidence. Thus on
pp. 130-131 he remarks on the existence of a so-called ‘nominativus
absolutus,’ but he does not tell the reader how frequent the phenomenon
is, and which Hebrew constructions were rendered by it. Likewise,
Thiersch has fine insights into the causative verbs of the LXX ending in
-d¢w, (Cw, -dw, and -bvw, of which -Ow and -((w usually reflect the
hiphal, but again, these remarks are impressionistic, very brief and
without reference to the frequency of the different renderings (pp. 151-
153). On the equivalence of the tenses of the Hebrew and Greek verb one
finds some remarks on pp. 153 ff.
Frankel’s study of translation technique is a real Vorarbeit, paving
the way to his subsequent discussion of the books of the Torah
(Einfluss). At the beginning of the chapter devoted to translation
technique (pp. 134-163), Frankel states that he wants to analyze how
the translators conceived of the individual parts of the Hebrew
language in order to better understand the ‘Ubersetzungsweise.’ The
discussion itself is subdivided into sections on the noun, verb, and
particles. The remarks themselves are very short and therefore of
limited value. Thus the first remark applies to the rendering of
construct combinations in which the second noun has been rendered with
an adjective rather than a noun, e.g. Gen 3:21 11 nind - xtTwvas
Seppattvous. In this case, Frankel provides just two examples, limiting
himself to the remark that this type of rendering occurs often in the
LXX (p. 134). However, what interests us more is how often this type of
rendering occurs, with which nouns and in which books. Likewise, with
regard to the present tense of the verb, Frankel remarks that ‘Das
Prasens wird haufig fur Perfectum gesetzt, e.g. Gen 15:2 O72 ox -
héyeu 5€ ABpay’ (p. 141). It is, however, of great interest to know how
often, in which conditions, and in which books the historic present is
used in the LXX. Occasional and haphazard as these remarks by
TRANSLATION TECHNIQUE 243
Thiersch and Frankel are, they advanced the study of the translation
technique to a great extent. For these two scholars determined some of
the categories which were to be studied in subsequent years, and
Thiersch’s insistence on the Hebraismi as the background of the
language of the LXX pointed to the direction which the research would
take. Somewhat more complete are the remarks by F.C. Conybeare and
St.G. Stock in the only full treatment to date of the syntax of the LXX
incorporated in their Selections from the Septuagint (Boston 1905). In
the treatment of syntax which precedes this work (pp. 50-97), much
attention is paid to the Hebraic background of many peculiarities of
the language of the LXX. This work, too, is brief, and it presents the
categories discussed by Thiersch and Frankel in a somewhat broader
fashion as a descriptive grammar with some background in the
translation technique. The work is aimed at the student who is trained
in classical Greek rather than the student of the LXX.
In the generations following Thiersch, the study of translation
technique was often incorporated in studies which analyzed the amount
of adherence of the translators to the Hebrew Vorlage. This aspect of
the language of the LXX intrigued scholars very much, probably
because of the background of these scholars themselves in New
Testament studies. Even the beginning student of New Testament Greek
realizes how much that language is indebted to the LXX, so that all
attempts to understand the Semitisms of the New Testament had to
start with the Hebraisms of the LXX. One of the earliest studies of this
kind was by Viteau, whose first major work was named Etude sur le grec
du Nouveau Testament. Le verbe, syntaxe des propositions (Paris 1893).
When Viteau realized in his conclusions (pp. 232-235) how much the
LXX influenced the New Testament, the title of his next work included
reference to the LXX.3 Likewise, the four-volume Grammar of New
Testament Greek, which was started by J.H. Moulton (I, 1906),
continued in collaboration with W.F. Howard (II, 1919-1929), and
completed by N. Turner (III, 1963; IV, 1976), contains much valuable
material on Semitisms in the New Testament and Hebraisms in the
LXX, including statistical evidence on the LXX. Many more New
Testament studies dwell at length on the LXX background of the
language of the New Testament. M. Johannessohn’s now classic study4
was written as a Vorarbeit for New Testament studies, as the author
3 Etude sur le grec du Nouveau Testament comparé avec celui des Septante. Sujet, complément et
attribut (Paris 1896).
“Das biblische kai egeneto und seine Geschichte,” Zeitschrift flir vergl. Sprachforschung 53
(1926) 161-212.
244 CHAPTER SIXTEEN
points out in his introductory remarks. Johannessohn wanted to show
how much the language of the New Testament owed to the LXX, and for
that purpose he investigated the Hebraic background of this phrase.
This has been further stressed by Dibelius in his review of
Johannessohn’s study: ‘Die Arbeit gehGrt in die Reihe der Einzelunter-
suchungen die heute allein imstande sind, das Problem der Septuaginta
Sprache und das der Abhangigkeit des urchristlichen Griechisch von
dieser Sprache der Lésung naher zu bringen.”°
In most of these works, the LXX and New Testament are rightly
discussed on a different level, but other scholars treat them on a par, as
if they are both components of one large so-called biblical language.
The indiscriminate discussion of the translation language of the LXX
and the language of the New Testament created many an imprecision.
The linguistic and lexical study of the LXX owes much to the study of
the New Testament. Serious students of the language and grammar of
the New Testament first analyze linguistic features and _ lexical
peculiarities of the New Testament from the LXX, and hence the
literature on the New Testament contains much relevant material on
the study of the LXXx.
While much of the interest in the language of the LXX derived from
studies of the New Testament, the LXX was also studied in its own
right. The Hebraistic nature of the language of the LXX remained one
of the main focuses of interest, so that even when the New Testament
was not explicitly mentioned, it was probably often interest in the New
Testament which directed this line of research. This becomes clear from
the introductory chapter in Thackeray, Grammar. A large section of the
Introduction is devoted to “The Semitic Element in LXX Greek’ (pp. 25—
55). The interest in Hebraisms is illustrated well by a thorough study
by R. Helbing, Kasussyntax, whose subtitle stresses its focus: Ein
Beitrag zur Hebraismenfrage und zur Syntax der ko.vy (Gottingen 1928).
This study analyzes in great detail the cases and prepositions used
with verbs in the LXX, and one of the main preoccupations of the author
is to show to what extent the translators were influenced by their
Hebrew Vorlage. Especially in such minutiae as case endings and
prepositions the real nature of the translation comes to light. Helbing
showed how in the wake of the Hebrew, new constructions were born in
the LXX which previously were unknown in Greek: tetotO€vat with
ént, reflecting >y nv, instead of the genetive used in classical Greek
with that verb, Baovdevetv with énl reflecting ?y 72n, instead of the
2 Gnomon 3 (1927) 646.
TRANSLATION TECHNIQUE 245
genetive used in classical Greek with Baotdevetv, and duvupt with év,
reflecting -2 viwi, instead of the accusative or dative used im classical
Greek with épvupte.
Thorough as this study is, it provides only ‘Bausteine’ for the study
of Hebraisms (p. V). The book has no concluding chapter, and such a
chapter probably would have necessitated a second monograph. A
conclusion would have referred to the different behavior of the verbs,
the definition of Hebraisms in this context and the different number of
Hebraisms found in the books of the LXX.
A study like that by Helbing contributes to the analysis of
translation technique, since it illustrates the translators’ dependence
upon the Hebrew in such minutiae as prepositions. Also other studies
written after Helbing show the translators’ dependence upon the
Hebrew. It is probably true to say that one of the focuses of interest in
translation technique is exactly this dependence of the translators on
Hebrew. At the same time, also scholars who did not set out to
investigate this topic arrived at the same view. Thus, the conclusion
cannot be avoided that the grammatical categories of the Hebrew
influenced the translation to a great extent.
Beyond the general interest in the Hebraic background of the LXX, in
recent decades several studies have been written which show an
interest in the translation technique for its own sake. As a rule, such
studies collect and analyze the data, and by so doing these studies
provide a basis for conclusions on more general matters. In this way
various areas of translation technique have been covered. These studies
may be subdivided into the following areas: the verb, prepositions,
word-order, pronouns, syntax, word choices and the degree of
literalness. Updated bibhography is provided in TCU, 69-71 and in
Dogniez, Bibliography.
The first study to be written in recent decades was by A. Wifstrand.®
In secular Greek the enclitic personal pronouns mostly precede the verb,
but in the LXX they usually come after the verb because of the Hebrew.
In Hebrew the pronouns are suffixed to the verb, e.g. yiqt®leni, or they
occur after the verb, and the translators simply followed this sequence.
Wifstrand investigated the different approaches of the various
translators to this matter. Of these, the most literal ones reflect the
grammatical habits of the Hebrew, while the free ones allow
themselves to place the pronouns before the verb in accordance with the
rules of the Greek language.
6 A. Wifstrand, “Die Stellung der enklitischen Personalpronomina bei den Septuaginta,”
Bulletin de la Société Royale des Lettres de Lund (Lund 1950) 44-70.
246 CHAPTER SIXTEEN
By far the greatest contribution to the study of translation technique by
a single scholar is found in the work of I. Soisalon-Soininen and his
students. Some of the topics covered by him are: the infinitives of the
LXX, treated in a 200-page book; and furthermore articles on the status
constructus, the Hebrew relative clause, the verb €ye.v, some types of
renderings of the preposition -2, the comparative jn, the genetivus
absolutus, and the independent personal pronouns. These studies are
collected in his Studien zur Septuaginta-Syntax (AASF B 237; Helsinki
1987). The work of Soisalon-Soininen is based on an extensive card
system in which the phenomena have been recorded. Two of Soisalon-
Soininen’s students wrote monographs on additional subjects.” These two
books as well as Soisalon-Soininen’s monograph on the infinitives,
provide important statistical data for the different books of the LXX.
7 R. Sollamo, Semtprepositions; Aejmelaeus, Parataxis (see p. 232, n. 14); eadem, On the
Trail of the Septuagint Translators—Collected Essays (Kampen 1993).
CHAPTER SEVENTEEN
RENDERINGS OF COMBINATIONS OF THE INFINITIVE
ABSOLUTE AND FINITE VERBS IN THE SEPTUAGINT—THEIR
NATURE AND DISTRIBUTION
1. Background
This study deals with the LXX equivalents of all occurrences of the
infinitive absolute in the Hebrew Bible which are combined with
finite verbal forms, denoted here as ‘infinitive absolute constructions.’
The translations of these constructions have been studied by Rieder
(1884), Hauschild (1893),! and Thackeray (1908),4 and are now
examined in detail with the aid of the Gdttingen editions and the
CATSS database,’ focusing on the distribution of the various types of
renderings. The statistics in the charts are intended to be exhaustive,
but the examples are not.
The following constructions are used in the LXX for the infinitive
absolute construction of the Hebrew.
1. An exact Greek equivalent of the construction qatol gatalty, that
is, a combination of an infinitive absolute and a finite verb, appears
only in two verses in the LXX:
Josh 17:13 wenAR? wan - e€orcApetoar S€ abtove otk EEwre-
Opevoav
1 A. Rieder, “Quae ad syntaxin Hebraicam, qua infinitivus absolutus cum verbo finito
eiusdem radicis coniungitur, planiorem faciendam ex lingua Graeca et Latina afferantur,”
Programm des Konigl. Friedrichsgymnasiums zu Gumbinnen (Gumbinnen 1884) 1-3; G.R.
Hauschild, “Die Verbindung finiter und infiniter Verbalformen desselben Stammes,”
Berichte des freien Deutschen Hochstiftes zu Frankfurt am Main NF 9,2 (1893) 99-126 (also
published separately [Frankfurt am Main 1893]). Rieder, who also studied the infinitive
absolute constructions in the Hebrew Bible (Leipzig 1872), treated its Greek renderings only
briefly. Hauschild dealt more extensively with the Greek and Latin reflections of the
infinitive absolute construction, together with the figura etymologica, focusing on the Greek
and Latin language and not on translation technique.
2 HStJ. Thackeray, “Renderings of the Infinitive Absolute in the Septuagint,” JTS 9
(1908) 597- 601; id., Grammar, 47-50.
See Tov, “Computerized Database.”
248 CHAPTER SEVENTEEN
Jer 44(51):25 nrwyn avy)... dyrPn OPA ... WY) Hwy - ToLotcaL TroLhoopev
éppelvaoal évepetvate ... Kal Tovodoat étrothoate
2. Of the various equivalents of the Hebrew construction,’ the
rendering which probably comes closest to the implication of the
Hebrew is that which takes the Hebrew as a strengthened expression
of the finite verb, reflecting that idea by an adverb. Thus the
traditional understanding of, e.g., Gen 15:13 ytn v7 is ‘you shall surely
know,’ and of Gen 40:15 "nan 231 ‘I was indeed stolen.’ The adverb used in
Greek may or may not be derived from the same root as the verb.°
Gen 32:13 Joy oR ww - Si Kaddig eb ce TroLhow
Exod 15:1 MT =- = EvddEwS yap 6eS6Eaotar
Num 22:17 772DN 72D—i«- ~Ss Ev TL LG yap TLULow GE
1 Sam 20:7 mmm - oaKdrnpiic dtroxpr6y
This type of rendering is rare (see n. 5). The two main renderings used in
the LXX are listed below as (3) and (4), occurring with differing
frequencies in the various books.
3. A frequently used type of translation renders the infinitive
absolute construction with a finite verb together with a Greek noun,
either in the dative or, less frequently, in the accusative case, in
singular or (rarely) in plural. In this way the close relationship
between the verbal forms gato! and gatalty is transferred to a close
relationship between a Greek finite verb and a noun, usually cognate.
Constructions such as these existed already in classical Greek (e.g. guy7j
devyw, yYauUW/yduov yapéw).
Gen 40:15 "N23. 333 - «=~ KNOT EKAGTMV
Gen 43:3 TWIT - Sayaptupta Stapepaptuprytar
Exod 22:15 mai a —- — hepv ty hepvret avy
Exod 23:22 youn yin’ - dKorj dkovonte (frequently)
Or in plural, for contextual reasons:
Lev 19:20 MMT X?2 TIM =«=—- ~—s Kal... AUTpoLg ob AEAUTPWTAL
Lev 24:16 mv our - AlBore ALBoBoAEL TW (also Exod
19513721228)
Num 23:25 WIPNRX?I IAP = =- ~—s Ka Tdpate KaTapdon ... adTdv
Likewise, with the accusative:®
4 For the use and meaning of the infinitive absolute constructions in biblical Hebrew, see
Gesenius—Kautzsch, Grammar, 342-345; A.B. Davidson, Hebrew Syntax (3d ed.; Edinburgh
ee) 116-120; see further earlier studies mentioned by Gesenius—Kautzsch.
° The full evidence, referred to in the chart, involves the following verses: Gen 32:13;
37:33; 46:4; Exod 15:1, 21; Num 22:17; 1 Sam 20:7; Jer 25:30; Am 9:8; Prov 23:1, 24; 27:23.
6 The full evidence, referred to in the chart, involves the following verses: Gen 19:9;
50:15; Exod 21:22; 22:25; Lev 5:19; 7:24; Num 23:11; 27:7; Deut 14:22; 15:10, 14; Josh 24:10;
INFINITIVE ABSOLUTE 249
Exod 22:25 2ann 73n
Num 23:11 a ASA2
Num 27:7 ynn na
Or with a preposition (only once):
Lev 7:24(14) WINN NP FINI
€ ve XUPAGUG €ve YUPAOTNG
evAOYNKaG EvAOYlav
Sd6ua Sue
Kal el¢ Bpaiotv ot BowhjceTtat
(cf. the Greek context)
Most renderings of this type use cognate nouns, as illustrated by the
above examples. Sometimes this relation is expressed by a noun derived
from a supplementary root, such as Bow and day:
Gen 2:16 DINN 72N = - ~— Bpuicet HayN
Gen 3:4 nanny - §avato dtobavetobe
In other cases cognate nouns differ slightly from the finite verb,
representing compounds or related words.”
Exod 19:13 por ipo = - «Ss EV... MLBOLG ALBOBoOANOSYGeTAL
Exod 21:20 opr dp. - lk €xSiKnO4Tw
Num 11:15 WNIT - — ATIOKTELVOV LE AVaLpeceL
Num 35:26 ReoNET - | €E68w EEEASH
1 Sam 12:25 wanyit - Kakla KaKoTOLHonTeE
4. The other main type of rendering represents the close relation-
ship between qatol and gatalty by a combination of the Greek
participle and finite verb, probably as close as the translators could
come within the possibilities of the Greek verbal system. In some cases
a different vocalization of the Hebrew may be surmised, e.g.
Gen 15:13 ylvaoKuwv yvoor possibly reflects y1n y7° instead of YT"
van
Cntuv e€eCyjmoev possibly reflects w17 wit instead of
wat wT
Lev 10:16
However, 1n most cases no different vocalization should be assumed,
since the consonantal pattern would be different. E.g.,
Gen 18:10 Dw Dw - étravaotpéedwv héw
Gen 22:17 2s [aa - evdoydv ELVAD'YT|OW OE
Num 30:7 man - YEVOLLEVN yévntaL
Therefore, the only explanation of these renderings is that the
translators considered them an adequate representation of the Hebrew
Judg 20:43A; Isa 22:17; Jer 31:18; 48:2; 50:34; 51:56; Nah 3:13. Twice a different noun is used
in the accusative: Gen 37:33; Jer 25:30.
The full evidence, alluded to in the chart, refers to these verses: Exod 17:14; 18:18;
19:13 (250); 21:28; 22:18, 22) Nun 1 115;35:26.
250 CHAPTER SEVENTEEN
construction. This type of rendering was used for both active and
passive verbal forms. Examples of the latter are:
2 Sam 17:10 on? ON = - «= THKOWEV TaKy\ceTaL
Jer 10:5 Niwa xii - alpdpeva apdyoovtat
Nah 3:13 inndimn - dvorydpevat dvorxy6yoovtat
In these constructions usually the cognate participle of the same root is
used, as in the above examples. Sometimes the participle represents a
supplementary root, such as \ey- and elt-:
Gen 18:18 mr yo - yivdyevog €otar
Judg 15:2 B nN TX = - hey el ta
Joel 2:26 Dix onraN) - Kal ddyeabe EaOlovtec
Sometimes? a participle of a synonymous verb is used.”
Gen 18:10 TWRIw - = éTavaoTtpéduv HEw
Exod 21:5 Tne ax = - =~ atop Bele elt
Exod 22:22 PUN’ PYRE - KEekpdEavtec kaTaBorowoL
Exod 23:4 yawn wi - droaotpépac droducerc
Lev 13:7 nwonnws - wpetaBadotica petatéon
Ps 109(108):10 wiry - cadevdpevol LETAVaOTHTWoaV
At times a compositum is used.
Lev 10:16 wrTwIT- CqTay €€eCthoev
Num 12:14 Prep yr - MTV évéTTvdEV
A variant of this type of rendering uses forms of elu or ylyvoyat with
adjectives.
Exod 22:12 mopy - Onptddwtov yévnta.
Num 22:38 Dow 3D --~S BvaTdcg Eoopar
Isa 40:30 Ywoa dwa - dvlayveg Ecovtat
8 Thackeray’s remark (p. 599) ‘... where this is used in the Pentateuch an attempt ts
often made to render it more classical by varying the verb’ is imprecise (see the numbers in
the chart). This applies also to Thackeray’s statement (ibid.) that this habit has been
abandoned in the books of the Kingdoms, although the numbers are not large (1 Sam 2:16;
20:3; 2 Kgs 14:10).
? In addition to the mentioned instances, see also Lev 14:48; Judg 16:11B; 1 Sam 2:16;
20:3; 2 Kgs 14:10; Jer 37:9; Hab 2:3; Ps 118:13.
INFINITIVE ABSOLUTE 251
In these cases it seems as if the two Greek words together represent only
one of the two Hebrew words, but the very use of two different Greek
words reflects the translator’s wish to represent the two words of the
Hebrew.
Almost exclusively the participle of the present tense (active,
passive, Or medium) is used (see the above examples), and for
exceptions see the aforementioned Exod 22:2; 23:4 as well as some other
instances in which the translator stressed the past aspect expressed by
the participle:
1 Sam 14:28 yawivawy - dpKkloac dpkicev
1 Sam 14:43 “nayo OVo = - = yevodpevog Eyevodunv
5. Yet a different way of expressing the infinitive absolute is the use
of adjectives. This occurs very rarely in the LXX.!°
Num 13:30 221 72> - =~ Buvatol Suvnodpeba
Amos 7:11 Mita - alypddwtog dxéhoetat
6. Combinations of gatol gatalty are often rendered by a Greek finite
verb only, as if the translator gave up an attempt to find a suitable
equivalent for the two words of the Hebrew. In some of these cases the
translator may have known a shorter Vorlage, but in most cases
different translation techniques must be presumed. The relatively large
number of such renderings in Isaiah (see below) probably points in this
direction.!!
Gen 27:30 NONE WNW - Kal €Ey€veto we €€jev
Gen 30:16 Tmiw wv > - pep loOwpat ydp oe
Gen 43:7 YT VITA - aT) HBELpev
Isa 24:20 vunyvi - €KALVE
Isa 56:3 YIy 77321 - ~— doptet pe dpa
10 The full evidence, referred to in the chart, involves the following verses: Gen 44:28;
Num 13:30; Amos 7:11,17; Job 14:18 (sub ast.).
1] The full evidence, referred to in the chart, involves the following verses: Gen 8:7;
20:7; 24:5; 27:30; 30:16; 31:30; 43:7,20; Exod 2:19; 5:23; 12:9; 21:19,36; 22:2.4,5,11,13; 23:5; Lev
10:18; 13:22; 20:13; 27:19; Num 21:2; 22:37; 24:11; Josh 6:13; 9:24; 23:12,13; Judg 11:25A,35A;
14:12A; 1 Sam 2:30 (= 4QSam?); 20:28; 23:22; 27:1; 2 Sam 3:24; 2 Kgs 5:11; Isa 10:16; 22:7;
24:20; 35:2; 36:15; 50:2,2; 55:2; 56:3; 59:11; Jer 11:7,12; 22:4; 42:15; 49:12; Ezek 1:3; 25:12; 31:11;
Am 3:5; Zech 8:21; Ps 50:21; Prov 23:5; Lam 3:20; Est 4:14.
252 CHAPTER SEVENTEEN
7. Translations which do not express the special meaning of the
Hebrew construction are rare:!?
Gen 26:11 nar nin - Bavdtw (8avdtov) €voxyog €otar. The almost
universal LXX equivalent, Savdtw with a verb of killing or dying, also
employed in Gen 2:17; 3:4, was not used here.
Job 13:10 oomx may nin = 6- ~~ oobbyv Artov dé yEet bude
On the basis of the aforementioned techniques we can sometimes recon-
struct from the LXX infinitive absolute constructions not found in MT.}8
Gen 19:17 wo) ?y ond = - =~ aie onife TH ceauTod wuyhv
= war 2y von von?
Num 30:6 NI - = dvavevuv dvavevan
= NY Nn (thus SP)
Jer 3:1 Nw - sot) dvaxduttovoa dvakduser
= Ww: Ww?
Jer 31(38):33 “min nAx enn = =—- ~— 8 Bote Sow vd_Louc Lou
= "nn NX "nn yn!
2
The distribution of the various types of renderings in the books of the
LXX is indicated in the chart with the following abbreviations:
1. Finite verb with participle
Pp finite verb with participle
pd idem, with different verb
pc idem, with compositum
2. aj finite verb with adjective
Sy = finite verb only
4. Vv varia
5. Finite verb with noun
nd cognate noun in dative
ndd idem, different noun
na cognate noun in accusative
nad idem, different noun
6. ad finite verb with adverb
12 The full evidence, referred to in the chart, involves the following verses: Gen 26:11;
Exod 22:3, 12; 34:7; Num 22:38; Josh 7:7; 17:13; 2 Kgs 3:23; Isa 40:30; Jer 44:25 (3 x); Ezek
33:16; Hos 10:15; Job 13:5,10, 17; 21:2; Dan 11:10.
3 The full evidence, referred to in the chart, involves the following verses: Gen 19:17;
47:22; Exod 11:9; 22:19; 23:22; Lev 24:21; Num 30:6; 35:21; Deut 13:16; 15:10; 1 Sam 2:25 2 x
(= 4QSam*); 2 Sam 20:18; 2 Kgs 11:15; Isa 19:22; Jer 3:1; 7:4; 22:24; 31:33; 32:28; 34:2.
INFINITIVE ABSOLUTE 253
Notes to the chart:
1. Hebrew infinitive absolutes reconstructed from the LXX are
indicated in the chart as ‘(+1).’
2. The statistics apply only to instances recognized by the translators
as infinitives absolutes, disregarding textual problems:
Exod 21:19 ROW XSW WM wnaw pr - ~— TrAY THe dpylac atrrod
dtoteloe. kal Ta Latpela
Josh 7:7 Taytmay1 - &eBlBacev 6 tate cou (T2Y)
Job 13:17 yinw naw - dkovoaTe dkovoaTe (ivnw WHY?)
3. Infinite absolute constructions lacking in the LXX (e.g., Josh 6:13)
are not included in the chart.
Book pl4 pd pce aj SY nd na nad ndd ad
Genesis 11(4+1) 1 1 1 8 1 12(+1) 2 1 2
Exodus 2 3 ll 3 17(+2) 2 8 2
Leviticus 5 2 1 4 ZoCtl) <2
Numbers 4 2 1 3 11(+1) 2 ah 1
Deut 9(+1) 21(+2) 2(+1)
Joshua 1 3 1 1
Judges-A 9 3 10
Judges-B 13 i 2 6
1Samuel 26(+2) 2 4 4 1 1
2Samuel = 13(+1) 1 6
1 Kings 11 4
2 Kings 2(+1) I 1 1 6(+1)
Isaiah 2 10 9(+1) 1
Jeremiah 25(+3) 1 5 12(+1) 4 1 (+1)
Ezekiel 2 3 1 18
Min Pr 12 1 1 2 2 i) 2 1 st 1
Psalms 5 2 1 1
Job 1 4 1
Proverbs 1 3
Lam 4 1
Ruth 1
Esther J 1
Canticles 1
Neh 1
1 Chron 4
2 Chron 2
14 This group includes supplementary forms, such as yw and eltov.
254 CHAPTER SEVENTEEN
Book Pp pd pe 4@j — Vv nd na nad ndd ad
Daniel d 1
Sirach 1 1 Z 1
Conclusions:
1. The two main types of renderings use either cognate participles (p,
pc) or nouns in the dative or accusative (nd, ndd, na, nad). Some books
display a relatively large number of renderings reflecting only finite
forms of the verb (-).
2. Some books reflect a distinct preference for a certain construc-tion.
The following books prefer the noun constructions: Exodus, Leviticus,
Numbers, Deuteronomy, Judges A, Ezekiel. The participle construction
is preferred in Judges B, 1-2 Samuel, 1 Kings, Jeremiah, and the Minor
Prophets. In other books the statistics are not conclusive. The one book
in which the rendering by the finite verb only occurs frequently is
Isaiah (see below). It is hard to know whether the distinction between
the two major types of renderings is that between the early and late
translation units. For example, in Genesis, probably the earliest
translation, they are equally distributed. Likewise, it is difficult to
know whether the different types of translation reflect a different
approach towards the translation technique. Since the participle
construction seems to be a more literal reflection of the Hebrew than the
noun construction, it may have been more at home in literal translation
units. This assumption fits the difference between the A and B
manuscripts of Judges (see below), and probably also that of the other
books, but for several books there is too little evidence.
3. Greek passive verbal forms tend to be used together with the noun
construction rather than with passive participles, probably because the
translators found the use of the passive participle together with finite
forms too complex. However, as the use of passive verbal forms is
rather rare, it cannot be claimed that they determined the choice
between one of the two major systems of representation.
4. The statistics should be analyzed cautiously and cannot be taken
at face value. Contents must be taken into consideration, especially in
books containing a relatively large number of identical renderings. Thus
in 2 Kings containing 6 instances of the noun construction as against 3 of
the participle construction nevertheless the latter is more frequent,
15 Thackeray's statement (Grammar, 598) that ‘the construction with the noun is always
used in the Pent. where the verb is in the passive’ is indeed correct for the Pentateuch,
but not for the other books (see, e.g., 1 Sam 2:27; 2 Sam 17:11; Jer 3:1; Am 5:5; Ps 118:13).
INFINITIVE ABSOLUTE 255
since all instances of the former pertain to renderings of nian nv, et sim.,
and n-nnncn. These two Hebrew constructions constitute also the
majority (12) of the 18 noun constructions in Ezekiel.
5. Translators did not distinguish between the translation of phrases
of the type gatol gatalty and the reverse sequence gatalty gatol.
6. The A text of Judges, preferring the noun construction, differs from
the B text, otherwise known to be more literal, preferring the
participle construction, E.g.,
Judg 7:19 wpnapr A éyépoer tyerpev
B éyelpovtec tyyetpav
Judg 11:30 In pm =A tTapadsudoet tapadsiac
B &180t¢6 Bic
Judg 17:3 ‘nwIpA wip OCA aytaop tyylaoa
B aytdCouoa tyylaka
7. Only Exodus and Isaiah contain a significant number of finite
forms rendering the two-word phrase. These should be considered free
translation options. The only other conglomeration of unusual trans-
lation options is in Job, which contains more unconventional than
conventional renderings. This, too, indicates a free approach to the
translation.
Content analysis:
1. Since Greek has no exact equivalent for the infinitive absolute
construction, different translation options developed, of which two
have become firmly established. The first translators probably
established certain translation habits, while later translators learned
from earlier ones. In this regard one should note the distribution in
Genesis in which the two main constructions are used in an equally large
number of instances, while in the next books of the Torah the noun
construction prevails. Probably the translator of Genesis was still
searching for the right type of rendering for the Hebrew construction.
2. Every translation unit contains exceptions to the majority
rendering, although it is unclear under what conditions the majority
rendering was abandoned. One possible explanation for such exceptions
in the Torah would be that the participle construction was used when
no appropriate cognate or other noun was found. This explanation would
apply to such verbs as x12 and 177, but not for ?xw, 7n1, 977, etc. For the
Torah and the other books the main reason for the different translation
habits appears to be inconsistency.
256 CHAPTER SEVENTEEN
3. A different explanation pertains to translation units which prefer
the participle construction and in which noun constructions are often
found. Since the noun constructions are preferred in four of the books of
the Torah, it seems that the translational tradition was often
instrumental, especially for a few frequently occurring expressions. This
applies especially to the noun constructions rendering nyan nw et sim. (46
x in the Bible; no exceptions), and n°nn "nN, et sim. (6 times in the Bible;
no exceptions), as visible in 2 Kings, Ezekiel and 1 Samuel. In 1 Samuel
the noun construction forms the minority rendering (five instances, four
of which pertain to nian mn: 14:39,44; 22:16). Likewise, throughout the
LXX there is considerable consistency concerning the rendering of ynw
“nunw, etc. (dKor] akovuw, et sim.).
4. At the same time, one often encounters internal inconsistency
within one verse, context, or translation unit, such as in:
Lev 13:7 mvonmwp - wpetaBarotoa petatéon
Lev 13:27 mwonimwy - 8taxvoer dtaxénrar
Lev 13:35 mo nmwp - §Lraxtoer diaxénqtar
Judg 11:25 ammn..amx tom - opt Kpeloowy ef ob ... wh paxn
oOn?lon71ox..327 - €paxéoato ... f TodkeLwv
ETTOAE LNOEV
In the same chapters of Jeremiah combinations of the infinitive
absolute with finite forms of 7n1 are rendered both with a noun (32:4;
34:2 [not in MT]) and a participle (31:33 [not in MT]; 32:28 [not in MT];
38:3).
5. Infinitive absolute constructions involving a specific Hebrew root
are rendered differently in the LXX as a whole as well as within
individual translation units. This fact underlines our contention that
the preferences of the translators were more instrumental in
determining the different renderings than anything else. Thus
combinations with v7" are rendered with a noun in the dative (Jer
40:14), a participle (Gen 15:13; 1 Sam 20:3,9; 28:1; 1 Kgs 2:37,42; Jer
26:15; 42:19), an adverb (Prov 27:23) and with a finite verb only (Gen
43:7; Josh 23:13). Combinations with 7n1 are rendered with a noun in the
dative (Gen 47:22; Judg 11:30; Jer 32:4; 34:2) or the accusative (Num
27:7), a participle (Deut 15:10; Judg 8:25; 2 Sam 5:19; Jer 31:33; 32:28;
38:3) or with a finite verb only (Num 21:2). The consistently rendered
occurrences of a few combinations, such as mentioned in remark 3, are
the exception rather than the rule.
CHAPTER EIGHTEEN
THEOLOGICALLY MOTIVATED EXEGESIS EMBEDDED IN THE
SEPTUAGINT
Es ist eine dankbare, fruchtbare Aufgabe, solche Vorarbeiten zu
machen, damit endlich auch einmal eine langst ersehnte Theologie der
Septuaginta geschrieben werden kann. (J. Ziegler, Die Septuaginta. Erbe
und Auftrag, Festfortrag ... 1962, 28 = Sylloge [Gottingen 1971] 613)
1. Introduction
Every translation of the Bible, or of any work dealing with one or more
deities, is bound to contain theologically motivated exegesis. The
Homeric songs and hymns, for example, deal extensively with the
world of the gods. Thus, any translation of Homer must take positions in
matters relating to these gods. It is difficult to know whether such
exegesis should be termed ‘theologically motivated.’ In the case of the
Hebrew Bible, however, almost any individual translation option is
potentially a carrier of theologically motivated (henceforth: ‘theol-
ogical’) exegesis, because of the central place of the Hebrew Bible in
the religions based on it. It is, in fact, difficult to imagine a biblical
translation without theological exegesis. Scholars have frequently
tried to locate such exegesis in different translations, hoping to find,
among other things, certain developments from one translation to the
next. Among all ancient translations, the Palestinian targumim
probably provide the most data on embedded theological exegesis (see
Levine, Aramaic Version).
Like all other biblical versions, the LXX reflects theological
exegesis, but probably to a lesser degree than the Aramaic targumimn.
Yet for many reasons scholars have paid more attention to this aspect
in the LXX than to the same phenomenon in other versions.! For one
thing, the LXX reflects a biblical text that differs considerably from
1 See the bibliographical appendix to my article “Die Septuaginta in ihrem theologischen
und traditionsgeschichtlichen Verhaltnis zur hebrdischen Bibel,” in M. Klopfenstein and
others (eds.), Mitte der Schrift? (Bern 1987) 237-268; M. Harl and others, La Bible grecque des
Septante (Paris 1988) 254-259; Dogniez, Bibliography.
258 CHAPTER EIGHTEEN
MT, and knowledge of its (theological) exegesis could provide many
helpful insights into text-critical problems. Such a prospect alone
justifies an intensive study of this aspect of the LXX. But there is more
at stake. Scholars are very interested to see how the content, ideas, and
words of the Hebrew Bible were translated or paraphrased by
translators living in the radically different Hellenistic world of
Alexandria. For generations scholars have tried to detect and to
delineate what the transfer of texts from Hebrew into Greek involved.
How much and what type of exegesis can be detected in the translation?
And to what extent does the translation contain that special type of
exegesis which we name ‘theological exegesis’? One should always
bear in mind that theological exegesis reflects but one aspect of the
general area of exegesis, and that it can neither be separated from the
translators’ approach towards exegesis, nor from our approach to
translation in other matters. There is another plausible explanation for
the interest of scholars in the theological renderings of the Greek
translators, an explanation which may well point to the major reason
for this special interest in the LXX. In many respects the content,
vocabulary, and ideas of the New Testament continue the world of the
LXX. Moreover, the New Testament contains a plethora of theological
terms; that is, many of its words, even common ones, have deep
theological meanings. Many of these have been studied often by
generations of clergymen and scholars. These studies did not leave the
LXX untouched, since it was recognized that the special vocabulary of
the New Testament was created in the LXX. Therefore along with the
interest in the theology of the New Testament, a special interest
developed in the theological background of the LXX.
By way of contrast, I have always preferred to ascribe deviations of
the LXX from MT to factors other than theological Tendenz. However,
while it may be true that books which reflect an easily perceived
degree of theological exegesis form a minority within the LXX canon, it
is clear that there are such books. It is also clear that relatively minor
amounts of theological exegesis are found in almost all books of the
LXX, but more substantial quantities are found in Isaiah, Daniel, Job,
and Proverbs. Not only do these books reflect many theologoumena, but
most of the examples of theological exegesis in the LXX derive, in fact,
from them.
The Greek form of these books does not reflect a systematic
theological system, but individual theologoumena, to be exemplified
below. Therefore one should not isolate theological exegesis from other
aspects of the translation. The books of the LXX contain no special
THEOLOGICALLY MOTIVATED EXEGESIS 259
theological message vis-a-vis their Hebrew text, and their layer of
theological exegesis is only part of a much wider stratum of (very) free
exegesis in a variety of matters. Those who minimized the existence of
theological exegesis were probably negatively influenced by the
exaggerated remarks in the literature which discovered theologou-
mena in almost every word of the LXX.
One such exaggerated approach is visible in the work of the scholars
contributing to ThDNT, a work which contains also extensive
information on the vocabulary of the LXX. Since many words in the
New Testament were described as theological, often rightly so, this
perception was projected back to the LXX, mostly without justification.
The dangers of this approach have been pointed out at length by J.
Barr, The Semantics of Biblical Language (Oxford 1961).
Theological exegesis of the LXX may be defined as any theological
element added to the source text by the translation. A similar
definition applies to other forms of exegesis as well. See TCU, 45-46.
The exegetical layer of the LXX consists of elements added as well
as omitted. Most exegetical elements, however, are reflected in the
lexical choices themselves, which were influenced by the immediate
context and the conceptual world of the translators.
Among these exegetical elements, theological exegesis is quite
prominent. It may relate to the description of God and His acts, the
Messiah, the exile as well as the whole spectrum of religious
experiences.
A translation may also reflect the intellectual background of its
translator(s). This background may be partly reconstructed by the
recognition of ideas and knowledge reflected in the choice of terms or
methods of expression in the translation. Such ideas and knowledge are
both idiosyncratic and culturally conditioned. It is not easy, however,
to identify such elements and to distinguish between the two strands
(personal, cultural). This is all the more so in the case of the LXX, since
its Hebrew Vorlage is not sufficiently known.
2. Theological exegesis in the LXX
The fact that the LXX is a translation should guide every detail of our
analysis of the theological exegesis of the LXX. All elements in the
translation somehow relate to the Hebrew Bible: they either reflect
their Hebrew counterparts or they are additions to certain elements in
the Hebrew Bible. We do not turn to elements common to both, for these
provide no indication of the intellectual and religious world of the
260 CHAPTER EIGHTEEN
translators. Nor are we interested in elements in the LXX which
presumably reflect Hebrew readings different from MT, even though
they may bear on theological issues. We focus only on the (Greek)
elements which the translation has in excess of the Hebrew, or which
have been deleted from the Hebrew, for in these elements we can often
recognize the mind(s) of the translator(s) and the world in which they
lived.
We approach the LXX as a translation and not as a basic source,
despite the fact that the LXX was for some ancients a primary
authoritative source, viz., for its Hellenistic Jewish readers, some of
the authors of the New Testament, and some Church Fathers. These
readers of the LXX often read and understood the translation in a way
different from that likely to have been intended by the translators (cf.
Tov, “Dimensions”*), especially in matters bearing on theology. When
analyzing the theology of the LXX translation we must try to
disassociate ourselves from the layer of exegesis that has been
attached to the LXX by the above-mentioned groups.
Focusing on the theological tendencies of the translators, we have no
other source of information than the LXX itself. There are no external
sources bearing on this issue (e.g., inscriptions, descriptions such as could
have been included in the Epistle of Aristeas).
a. Individual equivalents
The LXX is a heterogeneous collection of translation units, so that no
homogeneous theological approach should be expected in it. That is,
there are no theological traits common to all the books of the Greek
canon. Only a few books reflect theological exegesis, and even they
differ one from another, so it is not surprising that no common
theological features can be discerned in the LXX. The only area in
which shared theological exegesis might be expected is that of the
common LXX vocabulary. For example, the almost universal rendering
of 7177 (the Lord) is kUptos, also used for *15N. If there is a theological
conception behind this word choice, it may have been shared by all the
translators. Likewise, if there is a theological motivation behind the
choice of 8eé¢ for 0°79x (God), of véyo¢s for 771n (Law) and of &:a64Kn for
n-12 (covenant), the translators would again share a theological
conception with respect to these word choices.
But this is not the case. Common renderings do not necessarily imply
a common stance on matters of exegesis. Some equations of Hebrew and
Greek words certainly carried theological overtones when they were
first introduced into the area of biblical translation—probably at first
THEOLOGICALLY MOTIVATED EXEGESIS 261
orally in Jewish-Greek circles—but this exegetical element was not
realized on each occasion when the word was used. For example,
although when the equivalents 717” - kUpto¢g and 171n - vépLos were first
used, they may have carried certain theological overtones. The first
translators seem quickly to have forgotten such implications, since they
often merely rendered Hebrew words or roots automatically with the
same Greek equivalents. The way in which some equivalents were used
almost exclusively for all occurrences of their Hebrew counterpart
makes it unlikely that the first translators were fully aware of the
semantic content of their renderings in each individual case. Many
examples of the automatic use of equivalents by the first translators,
that is, the translators of the Torah, can be given. It is even more the
case for the later translators who accepted the translation vocabulary
from their predecessors. The later translators depended on the
translation of the Torah to the extent that it even served them as a
dictionary (see Tov, “Pentateuch’*).
In point of fact, the presence of theological exegesis in standard
renderings is rare. The majority of translation equivalents derive from
linguistic identifications of a given Hebrew root or word with a Greek
equivalent; as such they are of more importance for our understanding of
the linguistic knowledge of the translators than for our understanding
of their conceptual world.
This is not a uniformly accepted view. Various scholars consider
even many of the standard renderings of the LXX to have been
theologically motivated. Long ago A. Deissmann claimed that ‘the
Bible whose God is Yahweh is a national Bible; the Bible whose God is
KUptos is a universal Bible.’ Similar to Deissmann’s view is that of
C.H. Dodd, The Bible and the Greeks (London 1935) who opines that
the renderings of m7 by ktptos, of O772x by Beds, and of n-72 by StabhKn
are determined by theological factors. Dodd makes similar claims both
for the standard rendering of 771n by véyos and for various other words
in the realm of religion (the names of God, words for righteousness,
mercy and truth, sin and atonement).
However, it is questionable whether many of the equivalents used
by the first translators (of the Torah) were indeed determined by
theological or by other tendencies. For one thing, it is frequently
difficult to distinguish between shades of meanings in the Hebrew and
Greek. In the case of n-12 (covenant) however this is not difficult. Did,
then, the standard rendering of n-72 by &taOtKn really reflect a certain
2 “Die Bibel deren Gott Yahveh heisst, ist die Bibel eines Volkes, die Bibel deren Gott
KupLos heisst, ist die Weltbibel,” Neue Jahrbiicher fiir das klassische Altertum 11 (1903) 174.
262 CHAPTER EIGHTEEN
theological view? In our view it did not. It simply reflected a certain
view of the meaning of n-72 in the Torah. See Tov, “Dimensions,”* for
details. The equivalent is exegetical; it also had theological
implications for the readers of the LXX, but for the translators
themselves it involved only semantic exegesis.
The identification of the root ?7x (to be righteous) with 8tkato- (e.g.
> 7x - 6lkatos) was accepted by most of the LXX translators because the
two words cover each other relatively well. At the same time, this
lexical choice has given rise to an interesting semantic development
within Greek: in classical Greek the stem 8tkato- is used mainly with
regard to the relationship between human beings, but in the LXX (as a
result of its Hebrew source) it refers predominantly to the relationship
between man and God. In our view, the special use of 6tkato- in the LXX
does not reflect any theological tendency of the translators, but it is an
inevitable result of the identification of the roots ?7x8 and 8tkato-. This
is an example of an equivalent in which many scholars have recognized
(theological) exegesis, or a shift in meanings, while in our view the
differences between the LXX and MT resulted from a process of lexical
identification coupled with the technique of stereotyped translation.
By the same token it is not likely that the translators understood a
different nuance in the meaning of Stkatogvvn which according to
Schreiner refers in classical Greek to ‘the virtue of human justice’ and
denotes ‘the mighty aid of God’ in the LXX.
A slightly more complicated case is the standard LXX translation of
11 with tpoorAutos. See Tov, “Dimensions.”* The equivalence 7) -
tmpoonrutos reflects the linguistic, rather than the theological back-
ground of the translators.
Similarly negative results are obtained from the analysis of the
equivalents 0°7?x - 866g (God), 012w - elprvn (peace), and wd: - puxt
(soul). J. Barr has demonstrated that also dydtm (love) in the LXX for
Manx is not ‘theologically motivated at all but has its basis in purely
linguistic features.”4
Most of the renderings reflect linguistic and semantic identifications,
which as a rule, did not imply further forms of exegesis, such as
theological exegesis. While the choice of these translation equivalents
3 J. Schreiner, “Hermeneutische Leitlinien in der Septuaginta,” in: O. Loretz und W.
Strolz (eds.), Schriften zur Weltgesprich 3. Die hermeneutische Frage in der Theologie (Freiburg
1968) 391.
4 “Words for Love in Biblical Greek,” in: L.D. Hurst and N.T. Wright (eds.), The Glory of
Christ in the New Testament, Studies in Christology in Memory of G.B. Caird (Oxford 1987) 3-18.
The quotation is from p. 5.
THEOLOGICALLY MOTIVATED EXEGESIS 263
certainly had theological implications for generations of LXX readers,
as a rule they did not have such implications for the translators
themselves. Thus, in our view @eés was a logical choice for 0°7x just as
KUpLos was for 717° (pronounced adonay) véuos for 171n, Sikatoovvn for
PTR, dyad for 127k, bux for w51, etc. Although in these equivalents, as
well as in others, theological exegesis has sometimes been postulated,
for many other stereotyped equivalents no such exegesis is found, e.g.,
ww - HALOS, IN - TaTp, ON - unTnp, Ow - dvoua. Thus some words have no
theological implications at all. Thus Seeligmann, Isaiah, 96-97:
The question is, to what extent the Greek terms employed were, in the
translator’s mind, charged with a significance which caused the
original biblical picture to be shifted to another plane of thought. In
attempting to answer this question, we should not, of course, assume
that the translator, in using these terms, was fully aware of their
etymological evolution or the development of their signification, both
of which have been determined only by modern science; neither are we
justified in assuming that he took full account of all the religio-historic
or religiophilosophic implications which might eventually be placed
upon his choice of words.
At the same time, some renderings do reflect theological exegesis.
a. The translator(s) of the Latter Prophets who rendered the phrase
MNXAR TW (literally: the Lord of armies) consistently with kUpLos
TavTokpdtwp (the Lord omnipotent) must have had a certain view of
the nature of the Hebrew phrase. For him (them) nixax included not just
a body of ‘angels’ or ‘armies,’ but it encompassed everything. Thus,
when choosing this rendering, the translator(s) exegetically rendered
the Hebrew, at the same time that he (they) opted for a term also
known from the world of the Greek gods, some of whom were described
as TavToKpdtwp. For details, see Tov, ”Dimensions.”*
B. The translators of the Torah made a consistent distinction between
a Jewish altar (N21 - 8SuotaoTHptov) and a pagan altar (Bwyuds); the
Aramaic targumim similarly distinguished between the Jewish xn27n
and the pagan x7X (viz., ‘heap’ of stones).° This distinction undoubted-
ly derived from the translators’ wish to differentiate the Jewish
religion from that of the non-Jews (8votaoTHpLov, a neologism in the
Greek language, was probably coined by the translators or at least
within Jewish Alexandria). Interestingly enough, the altar erected by
> For similar cases, see the distinction between idolatrous priests (]"7213) and priests of
God (87315, 77175), idols (xN1IYVD, Vo) and God (X779x, 1779xR). See L. Smolar and M.
Aberbach, Studies in Targum Jonathan (New York/Baltimore 1983) 154.
264 CHAPTER EIGHTEEN
Aaron in Exod 32:5 for the golden calf was not named Buwyds, but
QuoLtaoTpLov, as in the Aramaic targumim (xn27), probably so as to
minimize his sins, a tendency also reflected in the medieval
commentaries. Daniel, Recherches, who described and documented in
detail the different renderings of natn, also provided a few additional
examples of distinctions made by the translators between cultic and
profane uses of words: the cultic 12 (high place) as opposed to the non-
cultic nina (mountains); the cultic and non-cultic on? (bread); and 7M1n -
‘sacrifice’ as opposed to ‘gift.’
y. A much quoted example is that of dvouta (literally: lawlessness)
and dvopos (lawless). Néuos, the constant equivalent of 771n (law,
‘Law’), must have played a very important part in Jewish Alexandria.
It was only natural that the more one talked about the importance and
virtues of the véos, the more frequently negative aspects of life would
be described as opposed to the véuos. It was recognized long ago by
Flashar® that for the translator of Psalms dvopla (lack of vép0s) was
often used to designate various forms of transgressions and irreligious-
ness. Thus, according to this translator, all these transgressions
constitute sins against the voyos, the Law. A similar trend is visible in
Isa 57:4 where “pw yt (offspring of deceit) is rendered by otépyia dvopov
(offspring without vdépos).
The aforementioned analysis referred to possible theological exege-
sis observable in stereotyped renderings. The following discussion
centers upon theologoumena extant in the renderings of individual
words, omissions, and additions.
b. Addition of details pertaining to the religious background
a. The translations of the LXX and L of Esther fill in the religious
background of the book which is lacking in MT. Likewise, the long
Additions to that book, deriving from the same hand as the main
translation of the LXX and L, fill in various details in the story. Thus,
in MT, Esther is not concerned about dietary laws when she dines with
the king, but in Add C 27-28 she is extremely concerned about this issue
(LXX and L text). D 8 mentions God’s intervention and C 20 mentions the
temple. See the detailed analysis in Tov, “Esther,”* section iv 4.
8. Isa 5:13 nyt 72a ("ay 172) 72) is rendered as ’... because they do not
know the Lord.’
6 M. Flashar, “Exegetische Studien zum Septuagintapsalter,” ZAW 31 (1912) 81-116;
161-189; 241-268.
THEOLOGICALLY MOTIVATED EXEGESIS 265
y. This phenomenon is particularly frequent in Proverbs, as
recognized especially by Bertram and Gerleman.’ Proverbs contains a
combination of secular and religious wisdom, so that religious
interpretations are not foreign to the Hebrew book. However, the Greek
translation contains more religious interpretations than its Hebrew
source, and much of this layer has evidently been added by the
translator, e.g.:
Prov 3:18 WD PDN 7 Opn? (7 on py)
(She [wisdom] is a tree of life) to those who lay hold of
her; those who hold her fast are called happy (RSV).
Kal Tots €tTEpEeLSopévois Ett’ avUTHY WS éml KUpLOV dGadadns
... and she is secure for all those who support themsel-
ves on her, as on the Lord
In this translation wisdom is reinterpreted as God.
Prov 13:15 yn ym Dw 3Dw
Good sense wins favour.
(1) cvveots dyafh bl85worv xdapv, (2) Td 8 yvuosvar vopov
Stavolas éoTtiv ayabiis
Sound discretion gives favour, and to know the Law is
the part of a sound understanding.
It stands to reason that in this double translation, the free one mentioning vépos
(2) is original, while the literal one was added subsequently. This rendering,
possibly influenced by vépos copov in the previous verse, transfers secular
wisdom to the religious realm. It has also been added to the LXX of 9:10. Likewise,
T! often identifies nv (knowledge) with xn-21x (e.g., Isa 28:9; 40:14; Hos 6:6).8
Especially frequent in this translation is the reinterpretation of non-
religious words as religious terms. Often the doeBets, the ungodly, are
brought into the picture:
Prov 1:7 Wa O°27N TOW) WAN
Fools despise wisdom and instruction.
codlav 5€ kal tatde(av doeBeis éEovbevrjcovoiv
The ungodly despise wisdom and instruction.
7 G. Bertram, “Die religidse Umdeutung altorientalischer Lebensweisheit in der griechi-
schen Ubersetzung des Alten Testaments,” ZAW 54 (1936) 152-167; G. Gerleman, Studies in
the Septuagint (LUA NF 1,52,3; Lund 1956) 36-57. In spite of the general title of Bertram”’s
article, it deals almost exclusively with Proverbs.
8 For further examples and a discussion, see P. Churgin, Targum Jonathan to the Prophets
(Yale Oriental Series XIV) 122-123.
266 CHAPTER EIGHTEEN
3335 7)7j2 OW 07770)
... but disgrace exalts fools.
ol 8¢ doeBeits twoav atiulav
... but the ungodly exalted disgrace.
Likewise, doeBis is introduced in the translation as an equivalent of
ywn (wicked) passim in Proverbs (as well as in Job and Psalms), of °°0>
(fool) in 1:22, 32; 13:19, and of a few other words.
In this translator’s picture of the world, mankind is divided into
‘poor’ and ‘rich,’ ‘good’ and ‘bad,’ ‘just’ and ‘unjust,’ even more than in
the Hebrew source. The Greek translator loses no opportunity to add
such adjectives to the translation or to change existing ones.
Prov 16:7 INN O7W? PINK O wer 7997 47 M873
When a man’s ways please the Lord, he makes even his
enemies to be at peace with him (RSV).
15:28a Sextal rapa kuplw dSol dvOpuTwv dixalwv Sia S€ avtTav
Kal ol €y@pol lror ylvovtat
The ways of the righteous men are acceptable with the
Lord, and through them even enemies become friends.
19:22 DSW RS WA DIYs
... and a poor man is better than a liar.
kpeloowv S€ tTwxds SlKatos Tt, TAOvVOLOS PEevoTNS
... and a righteous poor man is better than a rich liar.
For similar additions of 8{katos in Proverbs, see 3:9; 10:17; 12:25 and for a change
along these lines, see 16:9 (LXX 15:29b).
c. Messianic interpretations.
Although there is not as much evidence for messianic interpretation in
the LXX as some scholars would like to believe,? the translation of
Numbers 24 does contain two instances of such exegesis:
Num 24:7 oan ona wan r>79 on 2r
Water shall flow from his buckets, and his seed shall
be in many waters (RSV).
? See the material analyzed by J. Lust, “Messianism and Septuagint,” VTSup 36 (1985)
174-191; “The Greek Verson of Balaam’s Third and Fourth Oracles. The dv@pwtros in Num
24:7 and 17. Messianism and Lexicography,” in Greenspoon—Munnich, VIII Congress, 233-
257; “Septuagint and Messianism, with a Special Emphasis on the Pentateuch,” in H. Graf
Reventlow (ed.), Theologische Probleme der Septuaginta und der hellenistischen Hermeneutik
(Projektgruppe Biblische Theologie, Tagungen in Pforzheim, 1993 und 1994; Gitersloh
1997) 26-45. Lust himself is very cautious; note, for example, the titles of the articles (not:
Messianism in the Septuagint).
THEOLOGICALLY MOTIVATED EXEGESIS 267
€Eedevoetat dvOpwtros €k Tod oTfé patos avtod Kal
Kuptevoelt E8vuv TOAAUY
A man shall come out of his seed, and he shall rule over
many nations.
Num 24:17 PNW) IW OI apy 2D 777
A star shall come forth out of Jacob, and a scepter shall
rise out of Israel (RSV).
dvatevet dotpov €E "laxw kal dvaothoetat dvOpwtos €&
‘lopanid
A star shall come forth out of Jacob, and a man shall
rise out of Israel.
The unusual word choices in both verses make it very probable that the Greek
translator, as well as the targumim, explained these verses as referring to the
Messiah. Likewise the translation of Gen 49:10 probably reflects such exegesis in
several unusual word choices.
d. Some theologoumena of the translator of Isaiah
a. The translator of Isaiah!9 frequently used 86Ea (glory), not only as
the standard translation of 312) (honor, glory), but also as an equivalent
of several other words, especially with reference to God (177, x), DX,
NNN, TY, -D°, 70n, 117). He even inserted it in the translation against
MT, e.g. Isa 6:1 23°77 nx ORD 1P7IW1, LXX: ‘and the house was full of his
glory (§6€a).’ Hence, for the translator of Isaiah, 56€a is one of the
central characteristics of God.!!
B. The idea that God brings owtt\ptov (salvation), referring parti-
cularly to salvation from the exile, has often been inserted into the LXX
against MT. E.g., Isa 38:11 o-nA PINI 7> 7? ANN xX? has been rendered as
‘I shall no more see at all the salvation of God...’. Isa 40:5 (9 7129 710)
927 ‘77D 7D IN? Awa 4D wn has been rendered as ‘and all flesh shall see
the salvation of God.’
e. Anti-anthropomorphic renderings
Although the translators generally felt free to render literally verses
or words in which God is portrayed anthropomorphically, in some
10 Fora description of this translator's theological inclinations, see Seeligmann, Isaiah, 95-
121 (‘The translation as a document of Jewish-Alexandrian theology’) and A. van der Kooij,
“Zur Theologie des Jesajabuches,” in: H.G. Reventlow (ed.), Theologische Probleme (see n. 9
above) 9-25.
11 Soe L.H. Brockington, “The Greek Translator of Isaiah and His Interest in 86€a,” VT 1
(1951) 23-32.
268 CHAPTER EIGHTEEN
instances anthropomorphic expressions were avoided. A few examples
of such renderings are: Num 12:8 o°2> ‘0 naam has been changed in the
translation to ‘and he beholds the &é€a (glory) of the Lord.’ Exod 4:24
‘1 wip has become ‘the angel of the Lord met him.’ The MT of Exod
24:10 says about Moses, Aaron, Nadab, Abihu and seventy of the elders
of Israel XW? 77?X NX Iw. The directness of this expression has been
toned down in the Greek translation to ‘and they saw the place where
the God of Israel stood.’. In Josh 9:14 1>xw x> ‘7-75 nx) the word ‘mouth’
has been omitted in the translation. The existence of such presumed
anti-anthropomorphic renderings has been emphasized by Fritsch,!4
but their number is actually much smaller than it would seem at the
outset, as shown by Orlinsky and his students in a series of
monographs.!3
Likewise, the LXX consistently avoided a literal translation of 1%
(rock) as an appellation of God. The most frequent equivalents are 8e6s
(God), Bon8ds (helper), PvAaE (guardian) and dvtiAjpTTwp (protec-
tor).!4
3. Limitations
The LXX translation reflects merely some theologoumena in a few
freely translated books as well as several theologically motivated
individual renderings, both the ones that occur occasionally and ones
that occur often in the LXX. Therefore, to speak of a Theologie der
Septuaginta (thus J. Ziegler, quoted in the beginning of this study) may
be not only an overstatement, but also an ideal that can never be
obtained, unless one defines the term ‘theology’ very loosely.
References to the ‘theology’ of the LXX that are not based on concrete
examples in given translation units, rather than in the LXX as a whole,
are therefore of very limited value. Hence, the many articles written,
for example, by Bertram did not do scholarship a good service. The
names of these articles are very impressive, referring to such matters as
“Zur Pragung der biblische Gottesvorstellung in der griechischen
Ubersetzung des Alten Testaments” (WO 2 [1954-1959] 502-513); “Vom
Wesen der Septuaginta-Frommigkeit” (WO 2 [1954-1959] 274~284); and
“Der religionsgeschichtliche Hintergrund der Erhohung in der Septua-
12 Ch.F. Fritsch, The Anti-anthropomorphisms of the Greek Pentateuch (Princeton 1943).
13 For bibliographical references, see Tov, “Die Septuaginta” (see n. 1) 257.
4 For details, see A. Wiegand, “Der Gottesname 118 und seine Deutung in dem Sinne
Bildner und Schépfer in den alten Jiidischen Literatur,” ZAW 10 (1890) 85-96; A. Passioni
dell’ Acqua, “La metafora biblica di Dio Roccia e la sua soppressione nelle antiche versioni,”
Ephem. Liturgicae 91 (1977) 417-453.
THEOLOGICALLY MOTIVATED EXEGESIS 269
ginta” (ZAW 68 [1956] 57-71; this study does not even mention any
specific LXX texts.). However, the content of none of these articles does
justice to the titles. Likewise, the section headings in J. Schreiner’s
study quoted in n. 3 (‘Zur Gottesvorstellung der Septuaginta,’ “Vom
Menschenbild der griechischen Bibel’) should have been phrased more
modestly, since there is insufficient information about these two
subjects. Schreiner jumps from one book to the other although the LXX is
a mere collection of books translated by different translators. The
background of each of these units has constantly to be taken into
consideration; any unit or series of units should never be taken as
representative for other translation units. For example, Schreiner, as
well as Bertram before him, draws conclusions from the rendering of the
divine name 77” as 6 Ltkavés in Ruth 1:20 ff. However, this translation
unit was rendered by kaige-Th and this reviser followed the midrash-
type exegesis of -71w as *1-w, ‘he who is sufficient.’ No conclusions should
be drawn for the LXX as a whole, nor even for the little book of Ruth.
Likewise, the LXX translation does not display signs of what Schreiner
(p. 375) calls in the wake of others ‘jiidisches Erwahlungsbewusstsein.’
Yet another example may be adduced to show how difficult it is to
draw conclusions about the tendencies reflected in the LXX. On the basis
of the following two verses, Bertram, ThWNT II, 643-4, s.v. Epyov,
claimed that ‘the negative attitude of Hellenistic Judaism to work
decisively affects the text’:
Gen 3:17 WIV ANT WIN
Cursed is the ground because of you.
ETLKATAPATOS 1 Yh EV TOLS Epyors Gov
Cursed is the ground in your labors.
Gen 8:21 DIN WIV TAIN DX TV 999 AON KX?
I will never again curse the ground because of man.
ob Tpocbyjow Ett TOU KaTapdcacbat Thy yr da Ta Epya
TiiV GVO TIWY
I will not curse the ground any more because of the
works of men.
However, the root 729 which underlies the translation reflects a Hebrew source
or one in the translator’s mind because of the context in both verses, where the
tilling of the ground is implied (3:17) or suggested itself to the translator (8:21).
For a similar rendering see Jer 14:4 777N7 Nayl - kal Ta Epya Tig yg.
If the difficulties described here are taken into consideration, and if
each translation is analyzed separately, several theological ideas can
nevertheless be isolated in the LXX, as described in section 2.
IV. THE SEPTUAGINT AND THE TEXTUAL CRITICISM
OF THE HEBREW BIBLE
CHAPTER NINETEEN
THE TEXTUAL AFFILIATIONS OF 4QSAMA
The importance of Samuel] scrolls from cave 4 has been recognized since
the first articles by F.M. Cross, in which two columns of 4QSam? and
seven fragments of 4QSam> have been published,’ and in their wake
many additional studies have written (see Dogniez, Bibliography). In
these studies, the Samuel scrolls have often been described as
‘Septuagintal,’ and the textual analysis of these scrolls has entailed
several textual theories. These theories were not limited to a
description of the main textual witnesses of Samuel (MT, LXX, and the
scrolls from cave 4), but they integrated the data in the description of
the relationship between the textual witnesses of the Bible as a whole,
and also in the reconstruction of its textual history. Because of the great
importance which is assigned to the Samuel scrolls, a major study of
4QSam? is reviewed here: E.Ch. Ulrich, The Qumran Text of Samuel
and Josephus (HSM 19; Missoula, MT 1978).
The title of the book under review does not reflect its contents well.
The book is not limited to a discussion of the relationship between the
Samuel scrolls and Josephus, but contains a full-scale analysis of the
textual affinities of 4QSam?. This textual analysis consists of a
detailed discussion of the relationship between 4QSam4? and the main
stream of the LXX (chapters II, IV), LXX&“< (chapter III), the MT of
Chronicles (chapter V), and Josephus’ biblical text (chapters VI-VIII).
The message of the book derives from an analysis of a few hundred
segments of text.
This analysis makes the book attractive, because these unpublished
readings of 4QSam4, often sections of two or three lines, add much to our
knowledge of this scroll. At the same time, however, the fragmentary
‘publication’ makes the evaluation of the discussion problematical.
1 “A New Qumran Fragment Related to the Original Hebrew Underlying the
Septuagint,” BASOR 132 (1953) 15-26 (4QSam‘*); “The Oldest Manuscripts from Qumran,”
JBL 74 (1955) 147-172 (4QSam”). Henceforth: Cross, 1953, and Cross, 1955. Some readings of
the Samuel scrolls have been incorporated into the textual notes to The New American Bible
(New York/London 1970), also published separately as Textual Notes on the New American
Bible (Paterson, N.J., n.d.), and further in BHS.
274 CHAPTER NINETEEN
Although there is @ priori no reason to doubt the correctness of the
author’s decisions with regard to his readings, his calculation of spaces
(which are often crucial to the discussion), and his joining of the
fragments, one has to rely on Ulrich’s judgment. At times, this situation
leads to some dissatisfaction, as the author’s text-critical analysis of
4QSam4 and the LXX is often based on palaeographical considerations
which cannot be evaluated. The very existence of a certain reading in
4QSam/ is often assumed on the basis of a single letter and, when that
letter is dubious, it would be helpful if the reader were able to consult
the photographs. For example, does 4QSam#@ in 1 Sam 2:22 read o1w[y>
(Cross, 1953) or o*wfiy (Ulrich, 73; LXX: étolovv, MT: wy)? Does
4QSam? in 2 Sam 3:8 read 712)X or wx]? This uncertainty refers also to
readings which are not analyzed in the book because they do not fit into
any of the patterns discussed. For example, does the exclusion from the
book of Jo yt in 1 Sam 2:3 indicate that the author read njivt = MT
(unlike Cross, 1953: Jny7), or that this instance was considered
irrelevant because of the difficulty in evaluating the LXX (yvuicewc)?
A second problem in the evaluation of the data discussed relates to
the author’s approach to the reconstruction of the Vorlage of the LXX.
The main interest of this book is the relationship between 4QSam? and
the LXX, and for this purpose the Vorlage of the LXX needs to be
reconstructed. After all, the text-critical analysis is based on Hebrew
readings common to the scrolls and the Vorlage of the LXX, and does not
concern their common exegesis. For this purpose, the author has
reconstructed a few hundred Hebrew words in the Vorlage of the LXX,
which are compared with 4QSam@. It would have been in order if the
author had outlined his view concerning retroverting in general and the
reconstruction of the Vorlage of the LXX of Samuel in particular. The
main requisite for such a reconstruction is one’s understanding of the
translation technique of the unit under investigation. There is obviously
a difference in this regard between the different sections of Samuel,
since one of its sections contains a literal translation (2 Samuel 10—1
Kings 2:11, ascribed to kaige-Th), while the other sections, the OG, are
neither very literal nor very free. The author has not, however,
indicated whether these differences in translation character affect his
approach to the reconstruction of their Vorlagen. Thus, are we entitled
to retrovert év kuplw ... €v Oew prov in 1 Sam 2:1 as °"79N2... "72 (against
MT ‘72... 712 = 4QSam? 7[w)-2 ... [717°2]) or should we ascribe the use of
two different Greek words to the translator’s wish to vary the
rendering of identical words in the same context? Likewise, does qv
AetToupya@v in 1 Sam 2:18 reflect nowy 77 (as in v. 11 MT and LXX), or
TEXTUAL AFFILIATIONS OF 40SAMA 275
only nqwn, as in MT of v. 18? Note also the difficulty in evaluating
yuuoews for MT my7 (1 Sam 2:3), mentioned above.
The reader also needs some guidance with regard to the orthography
of the reconstructed Vorlage which is represented by Ulrich sometimes
as the orthography of MT and at other times as that of 4QSamé? (e.¢.,
mo1?w on p. 45 and 1°17 passim). Likewise, how can one decide on such
minutiae as the Vorlage of €ya ... €yu ... €yod in 2 Sam 2:23-24 as ... "18
2 Nee “anstead Of “SiN) 0 “SIN. 518 in MT arid 738 ...°D1k a in
4QSam? Furthermore, one notes that in chapter V the lacunae in
4QSam/? are reconstructed on the basis of the MT of Chronicles, while in
the other chapters such lacunae are reconstructed on the basis of the
LXX of Samuel. On the whole, the lacunae in 4QSam? are more
frequently filled with words retroverted from the LXX than with
parallel elements in MT. Although the close relationship between
4AQSam/? and the LXX cannot be denied, it would have been better to
give as few as possible reconstructions for the lacunae in 4QSam?@.
Reconstructions should be limited to those instances in which one wants
to show that either MT or the reconstructed Vorlage of the LXX fits into
a certain lacuna and thus justifies the reconstruction of the whole line.
Despite the positive evidence for the close relationship between
4QSam? and the LXX, the two sources provide independent texts (see
below). 4QSam? should not be reconstructed or supplemented as if it
were the Hebrew text from which the LXX was translated.
These methodological problems occasionally affect the validity of
the argument. But while the retroversions from LXXL"¢ are often
problematical, those from the LXX are less so. Retroversions of
elements in the kaige-Th section are, as a rule, reliable, and this refers
also to most of the retroversions in the non-kaige sections mentioned by
Ulrich. Furthermore, the reconstruction of the Vorlage of the LXX of
Samuel has been made easier through the finding of the Qumran scrolls
themselves (see Tov, “Qumran”*, section 1 and TCU, 78-81). The central
themes of the book are summarized as following on p. 15:
‘The problem with which we are dealing falls basically under three
headings: Qumran, proto-Lucian and Josephus. We are interested to
know: (1) What new does the major Samuel scroll from Qumran tell us
about the ancient form of the text from Samuel? What are its kindred
text forms (e.g OG, pL, OL, C, J)? (2) What can we establish concerning
the early stratum of L? Since a 4Q texttype in Hebrew would have
been unavailable to Lucian, what can 4Q tell us about the proto-
Lucian text which formed the basis of the Lucianic recension? (3) Can
it be known which type of Samuel text J used? In which language?’
276 CHAPTER NINETEEN
The working hypothesis of this monograph is introduced on pp. 4-6: (1)
The ‘LXX’ of Samuel is heterogeneous since the manuscripts of the ‘LXX’
in 2 Samuel 10—1 Kings 2:11 contain kaige-Th, while the remainder of
the book reflects the OG. (2) LXXL"¢ in Samuel is composed of ‘at least
two strata’—a late stratum, designated Lz, dating from the time of the
historical Lucian, and an earlier, proto-Lucianic, stratum.
The main focus of the book is the relationship between 4QSam4 and
the LXX. On the basis of the consensus in modern scholarship the
author takes into consideration the possibility that 4QSam? relates
differently towards the kaige-Th and the non-kaige sections. Two
refinements in the presentation facilitate the evaluation of the
evidence: the agreements are divided into pluses, minuses, and
variants. Furthermore, a distinction is made between significant and
less significant examples, because often the nature of the examples is
more significant than their number.
Among the striking agreements between 4QSam@ and the LXX one
notes particularly pluses in 1 Sam 1:11 (based on considerations of space
in 4QSam?), 1 Sam 2:23, and 2 Sam 8:7-8. Of the significant minuses of
4Q0Samé? and the LXX, special mention should be made of 1 Sam 2:22 and
1 Sam 2:32a. Significant common variants may be exemplified by
1 Sam 1:24 4QSam? wow apa []2 1D2, LXX: év pdaxw tpret(Covt. =
wown 2 (MT: aw?w oD)
1 Sam 2:29 4QSam? wrrn 4°72n? (apparently also underlying the
LXX; see Ulrich), against MT m-wxnn o2N77277
2 Sam 7:23 4QSamé4 07978) = LXX kal oxnvijyata against MT 17x).
The large number of agreements between 4QSam@ and the LXX leaves
no doubt regarding the close relationship between the two sources. This
fact is significant not only for for the evaluation of 4QSam?, but also for
the text-critical use of the LXX in general and the book of Samuel in
particular. The importance ascribed to this translation by Wellhausen,
Samuel and Driver, Samuel, merely on the basis of intuition and
insight, is now supported by factual evidence.
Ulrich makes one further step. There are 124 instances of the pattern
4QSam4 = LXX # MT in the non-kaige section as against twenty such
examples in the kaige-Th section. Although the precise details of the
statistics are open to modification (see below), the relationship of the
LXX to 4QSam? differs from one section to the other. Taking into
consideration that the preserved fragments of 4QSam? in the non-kaige
section contain twice as many verses as in the kaige-Th section, 4QSam?
agrees three times more with the LXX in the non-kaige section than in
the kaige-Th section. These data provide helpful support for the
TEXTUAL AFFILIATIONS OF 4QSAMA4 277
theory that the kaige-Th section contains a revised Greek text.
Presumably the close agreement between 4QSam/@ and the LXX was kept
intact only in the non-kaige section which reflects the OG translation,
but disappeared in the kaige-Th section because it has been revised
towards MT.
Chapter III deals with the relationship between 4QSam? and
LXXLUuc, The discussion is limited to agreements between the two texts,
belonging mainly to the pattern 4QSam’4 = LXX/"¢ ¥ LXX, but also
4QSam@ = LXXLuC MT # LXX. The evidence is not as monumental as for
the pattern 4QSam? = LXX (chapter IT), but some important agreements
between 4QSam? and LXX4¥°¢ must be taken into consideration in the
evaluation of these sources. The agreement between 4QSamé@ ani and
LXXLUC tobe TeO6aloug in 1 Sam 5:9 (MT ink = LXX avTyVv) may serve as
an example.
As for the background of this analysis, the discussion of the Lucianic
problem has taken an important place ever since the pioneering work by
A. Rahlfs (for details, see Tov, “Lucian”*). Ulrich’s analysis should be
promising because he had access to all the material of 4QSam?, whose
close connections with LXX/¢ has not been investigated in full.
Ulrich makes a distinction between agreements between 4QSam? and
LXXLu¢ jn the kaige-Th section and in the non-kaige section. In the non-
kaige section eight items belong to the pattern 4QSam? = LXxluc # LXX
MT, while in the kaige section there are 27 such examples. Taking into
consideration that there is twice as much evidence available for
4QSam? in the non-kaige section as in the kaige-Th section, the agree-
ment between 4QSam? and LXX"< is seven times larger in the kaige-Th
section than in the non-kaige section. This evidence supports the
assumption of a distinction between kaige-Th and non-kaige. Further-
more, the new evidence provides independent proof for another
assumption which, in some way or other, has been accepted by modern
scholarship: in that section in which the manuscripts of the “LXX’
reflect kaige-Th, the OG has not been lost, but rather is somehow
reflected in the substratum of LXXLUC. If the text of LXX>"¢ agrees seven
times more with 4QSam/? in the kaige-Th section than in the non-kaige
section and if, furthermore, the pattern 4QSam?* = LXX has been
established firmly in the non-kaige section (chapter II), the two
conclusions can now be combined with reference to the whole book of
Samuel: the agreement of 4QSam? and the LXX in the non-kaige section
points to the OG translation and the agreement between 4QSam? and
LXX uc in the kaige-Th section apparently also refers to the OG
translation. Thus, throughout Samuel, 4QSam? agrees with the OG,
278 CHAPTER NINETEEN
even though this agreement has been obscured by the vicissitudes of the
transmission of the LXX. The fact that the agreement between 4QSam4
and LXXL"< in the kaige-Th section is not as extensive as between
4QSam? and the LXX in the non-kaige section must be ascribed to the
influence of the changes towards MT in LXXLUC by the historical
Lucian.
The following remarks should be added to this analysis:
a. When the agreements of 4QSam? with LXXLU¢ are set against
those of 4QSam? with the LXX, the examples of the former group are
not convincing. The relatively small amount of agreement between
4QSamé@ and LXX/"¢ must probably be ascribed to the changes inserted
by the historical Lucian. Whatever the reason, the agreements between
4QSam? and LXX"“ are not monumental. One should further take into
consideration that Ulrich focused on the agreements between 4QSam?4
and the LXX, while the disagreements between the two have been
disregarded (unlike the disagreements between 4QSam? and the LXX
discussed in chapter IV). When these disagreements are taken into
account, the pattern 4QSam4 = LXX!C is even weaker than it appears
now. In principle, agreements of the pattern 4QSam? = La, and/or
Josephus could strengthen the pattern 4QSam? = LXXL"¢, since all
sources involved may reflect the OG, but at the present stage of
research, such procedure should be considered a petitio principit.
b. It has been suggested by Cross, “Biblical Text” that the substratum
of LXX“"¢ contains an inner Greek revision towards a Hebrew text like
4QSam4. Ulrich supports this assumption by the eight examples of the
pattern 4QSam? = LXX/"° # LXX MT in the non-kaige section. However,
these few examples may also indicate that the OG basis of LXX/"¢ in
the non-kaige section differed slightly from the LXX in that section,
which presumably also contains the OG, or rather, an OG translation.
Alternatively, the LXX has been revised in these instances (and
elsewhere?) towards MT. In any event, the evidence does not prove that
proto-Lucian contained a revision.
c. The twenty examples of the pattern 4QSam? = LXX # MT in the
kaige-Th section apparently belonged to the OG substratum of the
Greek translation (pp. 92-93). These items can be combined with the 27
examples of the pattern 4QSam?@ = LXXLUC # LXX MT in the same
section because these, too, reflect the OG. On the basis of these
assumptions there are thus 47 agreements between 4QSam? and the OG
in the kaige section. This is an important datum, because these 47
examples can be compared with the 124 items of agreement belonging to
the pattern 4QSam? = LXX # MT in the non-kaige section (see above).
TEXTUAL AFFILIATIONS OF 40SAMA 279
Taking into account that the preserved verses of 4QSam#@ in the non-
kaige section are twice as numerous as in the kaige-Th section, one notes
that the agreement between 4QSam@ and the OG refers to 94 items (2 x
47) in the kaige section, and to 124 items in the non-kaige section.
Therefore the agreement of 4QSam? and the OG translation is more or
less equal in both sections of Samuel, a fact which corroborates the very
assumptions which have been mentioned above. This calculation has
not been made by Ulrich, probably because he considers the substratum
of LXXL¢ as reflecting a proto-Lucianic revision rather than the OG.
At this stage in the book the close agreement between 4QSam? and
the OG has been substantiated (in Ulrich’s words ’... that the Greek
version was originally translated from a Hebrew text much closer to
4QSam than to M’ [p. 119]). The next chapter (IV) is devoted to
evidence contradicting this theory. This evidence, belonging to the
pattern 4QSam? # LXX, is subdivided into four categories: 4QSam?4 = MT
+ LXX; 4QOSam? # MT # LXX; 4QSam? # MT LXX; 4QSam?4 = MT # LXX.
Expressing the relationship between different sources by way of
statistics is even more difficult in this chapter than elsewhere. For
some of the examples refer to complete sections differing in two or three
witnesses such as 1 Sam 2:13-16, while other examples refer to isolated
disagreements. There are four items of the pattern 4QSam? = MT + LXX,
but these examples actually belong to the pattern 4QSam? 4 LXX # MT.
Furthermore, there are four examples of composite differences of the
type 4QSam? # LXX # MT as well as 19 simple disagreements of this
type. There are also 23 items of the type 4QSam* # MT LXX and 18
items of the type 4QSam? = MT # LXX. Altogether, this chapter
contains 68 examples of disagreement between 4QSamé and the LXX, of
which several are composite. To this number we may add the 8 items of
disagreement between 4QSam? and LXX4UC in the kaige section.
However, most of these 76 disagreements are toned down by the
author, who describes them as secondary (error, omission, doublet), and
hence irrelevant. Thus, of the 76 items of the pattern 4QSam? # LXX,
the author accepts only 24 as valid. Ulrich then contrasts these 24
disagreements between 4QSam? and the LXX with the 183 items of
agreement between 4QSam? and the OG (the combined evidence of
chapters II and III referring to both the LXX and LXX"‘), concluding
that the differences between 4QSam? and the LXX can be disregarded.
There are differences, but the agreement between 4QSam4 and the LXX
is many times more convincing, concludes Ulrich.
Chapter V deals with ‘the agreement of 4QSam? with Chronicles’
against the MT of Samuel. Some examples of this pattern have been
280 CHAPTER NINETEEN
noticed earlier by Cross, “Biblical Text,” and Lemke.* The best
examples belong to the pattern 4QSam? = Chronicles # MT LXx,
especially in long pluses in 2 Sam 10:6-7 and 24:16. Ulrich also provides
a long list of minor agreements between 4QSam?, Chronicles and the
LXX of Samuel against the MT of Samuel. However the latter examples
can also be explained on the basis of the known pattern 4QSam? = LXX.
The author concludes that 4QSam4 and the MT of Chronicles reflect the
same Palestinian text which is distinct from the MT of Samuel.
Accordingly, one should not automatically ascribe the differences
between the MT of Samuel and Chronicles to the Chronicler’s editorial
activities, but apparently the Chronicler often used a Hebrew text that
differed from the MT of Samuel. This analysis illustrates just one
aspect of the relationship between 4QSamé? and the Chronicler, viz.
their agreements, disregarding disagreements between the two. These
disagreements cannot be evaluated easily because they may have
originated not only from differences between 4QSam and the
Chronicler’s Vorlage, but also from the Chronicler’s manipulations.
The author has shown that 4QSam4 is joined not only by the LXX
and LXXL¥< but also by the MT of Chronicles, and that their combined
readings often disagree with the MT of Samuel. This group is now joined
by a witness of different nature, viz. Josephus’ biblical text. The most
significant examples of the pattern 4QSam? = Jos. consist of a few
exclusive agreements between these two sources. E. g., the plus in 1 Sam
1:22 adw tv V1 wA[NN (about Samuel, cf. Ant. V, 347). Likewise, before
the beginning of the story in 1 Sam 11:1, both 4QSam? and Ant. VI, 68-
69 add a large section (see Ulrich and subsequently F.M. Cross °).
Except for the exclusive agreements of 4QSam? and Josephus, Ulrich
mentions long lists of other agreements (Jos. = 4QSam? LXX # MT; Jos. =
4QSam? LXXL< 4 MT LXX; Jos. = 4QSam? Chronicles # MT LXX; Jos.
4QSam2 MT LXX“< La # LXX).
Ulrich also mentions other examples for the same sources, but this
time when disagreeing with Josephus. Of particular interest is a group
of disagreements between 4QSamé@ and Josephus, when the latter is
based on a Greek rather than Hebrew source. For example, in Josephus’
quotation of 2 Sam 10:6, he speaks about ‘2X Upoc, the King of the
Mesopotamians.’ This text must have been based on a Greek reading
like LXXLUC (tov LUpov), which has been misunderstood as the name of
a person rather than a country (MT ox).
2 WE Lemke, “The Synoptic Problem in the Chronicler’s History,” HTAR 58 (1965) 349-
363. The conclusions of neither Cross nor Lemke are mentioned by Ulrich.
For details, see TCHB, 342-344.
TEXTUAL AFFILIATIONS OF 40SAMA 281
Summarizing both sections of this chapter, Ulrich concedes that
Josephus must have used a Greek text of Samuel which was ‘strikingly
close to 4QSam4, but ... that text was in the Greek language, closely
connected with OG\protoLucian and clearly distant from both M and
the kaige and hexaplaric recensions’ (p. 191).
Since Ulrich’s conclusions on Josephus’ biblical text have
implications for other topics, the last chapter (VIII), deals with
‘Josephus and his Vorlage..’ The author suggests that Josephus used a
Greek rather than a Hebrew text as the basis for his paraphrase of the
history of the biblical period. This assertion is based on an analysis of
statements in Ant. I, 5 and Ap. I, 54; I, 1. On the empirical level this
assumption is based on detailed analysis of Josephus’ paraphrase of 2
Samuel 6 and 1 Chronicles 11 in Ant. VII, 78-89. The Greek texts are
presented in parallel columns, while the degree of closeness between
Josephus and the Greek texts (LXX and LXX4¥°) of Samuel and
Chronicles is rated from 1-4 (rate 4 refers to the clearest evidence of
Josephus’ reliance upon an existing Greek text). The two most telling
examples are 2 Sam 6:8 x1yv p19 - &lakomy "O¢a = Ant. VII, 82 Oda
StakoTm; 2 Sam 6:19 NOX AWN INN TDWRI NAN an? Nn - KoAAUPLSa dpTou
kal €oxyap(tHv kal Adyavov amd THydvou = Ant. VII, 86 koAAuplda dptou
Kat €oxap( Thy kal Adyavov Thyaviotév. The Greek renderings are very
rare in the LXX, and, likewise, the words themselves are very rare (or
hapax legomena) within the Greek language, so that Josephus must
have relied on the wording of this passage in the LXX.
Ulrich also examines evidence which is contrary to his main
contention in this chapter, viz. evidence showing that Josephus’
biblical text is based on a Semitic Vorlage. Few positive arguments in
favor of this assumption can be sustained, so that we are left with the
conclusion that Josephus used a Greek text .
Chapter VII (‘Quantitative analysis of 2 Samuel 6’) provides a pilot
study of one chapter, analyzing Josephus’ biblical text in a running
passage rather than isolated verses. The following conclusions are
reached:
a. Of the three Hebrew texts, 4QSam4 is the best one, followed by
Chronicles. The MT of Samuel is a ‘poorly preserved text’ (p. 197).
b. The Greek translation of this chapter, though included in the OG
section, presents evidence of revisional activity (‘a later translator or
the kaige recension’). The evidence for this assertion is not convincing,
since the phenomena analyzed in this section were probably caused by
the vicissitudes of the textual transmission. Furthermore, the theory
suggested here is not a necessary link in Ulrich’s chain of arguments.
282 CHAPTER NINETEEN
c. 4QSam? agrees in eleven instances with the MT of Samuel against
the MT of Chronicles, but it also agrees in thirteen instances with
Chronicles against the MT of Samuel, and in eight instances it presents
independent evidence.
d. Ulrich discards the possibility that 4QSam? has been corrected on
the basis of Chronicles. The two texts are closely related, but they also
reflect independent developments occurring after the split of the two
texts from their common ancestor. In any event, the small differences in
details between the readings of the pattern 4QSam? = Chronicles #
Samuel (MT) preclude the assumption that 4QSam? was corrected on
the basis of Chronicles. This possibility seems unlikely also because of
the frequently occurring pattern 4QSam? = Chronicles = LXX Samuel #
MT Samuel.
The greater part of the author’s conclusions on pp. 257-259 is devoted
to an analysis of proto-Lucian. However, in the book itself this issue is
not covered at length, and when it is discussed it forms part of the
overall discussion of 4QSam?’s relationship to the Greek version of
Samuel. The main conclusions are: |
a. 41QSam? agrees significantly with the LXX of Samuel against MT.
b. Barthélemy’s theory with regard to the revision of the LXX of 2
Sam 10:1—1 Kgs 2:11 by kaige-Th is supported by the pattern 4QSam? =
LXX # MT which occurs much more frequently in the OG section of
Samuel than in the kaige-Th section, where the original agreements
with 4QSam? had been removed by kaige-Th.
c. The agreements of the type 4QSam? = LXXLUc # LXX MT are not as
monumental as with regard to the LXX.
d. The different ratio of agreement between 4QSam? and LXX/< (in
the pattern 4QSam? = LXxXbUc # LXX MT) in the different sections
(seven times more in the kaige-Th section than in the non-kaige section)
leads to the same distinction between kaige-Th and non-kaige as
referred to in conclusion b. Furthermore, the proto-Lucianic stratum of
LXxXbuc agrees with 4QSam/@ in the kaige-Th section, because in that
section proto-Lucian reflects the OG.
e. Eight examples belonging to the pattern 4QSam4 = LXXLuc # LXX
MT in the non-kaige section support the assumption that the substratum
of LXXL¥¢ contained a proto-Lucianic revision towards a Hebrew text
like 4QSam2.
f. In view of the 183 agreements between 4QSam? and the OG (LXX
and LXXLuCc), the disagreements between 4QSam4 and the LXX (76,
reduced by Ulrich to 24) may be disregarded.
TEXTUAL AFFILIATIONS OF 40SAM4 283
g. Significant data belong to the pattern 4QSam? = Chronicles # LXX
MT (both of Samuel). Late influence of Chronicles on 4QSam? is
excluded.
h. Josephus’ biblical text is based on a Greek rather than a Hebrew
text. This text is close to the text of 4Q0Sam4 and Chronicles, and the
LXX, LXX¥°, and Old Latin versions of Samuel, and differs from the
MT of that book.
i. The MT of Samuel is often corrupt, and differs from the
‘Palestinian’ Hebrew text of 4QSamé4 and Chronicles. This Palestinian
text formed the basis of the translations of the LXX, LXXL¥*, and La, as
well as for Josephus’ paraphrase.
Some reactions:
a. The relationship between the sources is determined on the basis of
statistics, but a refined methodology would have been in order. Thus,
Ulrich’s statistics do not distinguish between significant and
insignificant agreements, nor between simple agreements, such as the
addition or admission of a small element, and composite agreements
involving complete verses.
b. In the evaluation of the agreement between 4QSamé@ and the LXX,
their disagreements should be taken into consideration as well.
c. Readings of 4QSam4 and the LXX which reflect a presumably
original text, as opposed to a presumed corruption in MT should be
disregarded, or be given special status (e.g., corrupt readings in 1 Sam
1:24, 24; 2:22; 2 Sam 13:21, 22:39; 24:17). At the same time, common
corruptions of 4QSam4 and the LXX should be included in the analysis
(such as the doublet in 1 Sam 2:23-24 and the readings in 1 Sam 10:27—
Lisland in: 2 Sam 7:23).
d. Agreements between 4QSamé? and the LXX (as opposed to corrupt
readings in MT) should be excluded from the statistics if they are joined
by one or more ancient versions, as is true for several of the items
discussed.
e. When these remarks are taken into consideration, the agreement
between 4QSam@ and the LXX is smaller than suggested by Ulrich, and
the amount of disagreement is larger than indicated. At the same time,
the number of unique and independent readings of 4OSamé is larger than
suggested by Ulrich.
In spite of these criticisms, 4QSam?’s frequent agreement with the
LXX and Josephus’ biblical text has been demonstrated beyond doubt.
CHAPTER TWENTY
THE CONTRIBUTION OF THE QUMRAN SCROLLS TO THE
UNDERSTANDING OF THE SEPTUAGINT
The two areas of LXX studies on which the discoveries in the Judean
Desert have made an impact, are: (1) the credibility of the reconstruc-
tion of elements in the Vorlage of the LXX, and (2) the recognition of a
close relation between the LXX and specific Qumran scrolls.
This study is limited to the Qumran scrolls, since very few unique
agreements have been found between the LXX on the one hand and the
Hebrew texts from Masada, Murabba‘at and Nahal Hever on the other,
for, in general terms, these texts reflect MT.
The questions raised here were touched upon as early as 1959 by
H.M. Orlinsky,! and subsequently by R.W. Klein, Textual Criticism of
the Old Testament (Philadelphia 1974), as described by its two sub-
titles, one on the cover, ‘From the Septuagint to Qumran,’2 and one on
the title page, ‘The Septuagint after Qumran.’
1. Support of the Qumran scrolls to the credibility of retroverting
The Qumran scrolls provided the first massive support for the
correctness of an approach that has been an integral part of scholarship
for more than three centuries, namely, the reconstructing of details in
the Vorlage of the LXX by way of retroversion. Before the Qumran
discoveries no such external support was available for this procedure.
After all, before 1947 there was little if any external evidence in
support of the assumption that a given deviation from MT in the LXX
should be reconstructed into Hebrew rather than explained away as the
translator’s exegesis. The great masters in this area of reconstruction,
from Cappellus to Houbigant, and from Wellhausen to S.R. Driver,
lit. Orlinsky, “Qumran and the Present State of Old Testament Text Studies: The
ee piusent Text,” JBL 78 (1959) 26-33.
It is not uncommon to find evidence of textua] or editorial mishap in books on textual
criticism. Regarding this example, one may conclude, with support from the foreword by
G.M. Tucker, that the intended focus of the book is ‘The Septuagint after Qumran.’
286 CHAPTER TWENTY
operated with such tools as grammars, lexica, and concordances to the
Greek and Hebrew Bibles, but, actually, their major source of
inspiration was their intuition. Guided by this intuition, the above-
mentioned scholars, as well as others, suggested many a retroversion for
readings in the LXX which deviated from MT. In search of support for
these intuitive retroversions one cannot turn to the other ancient
translations, since these are equally as suspect of reflecting content
exegesis as the LXX. Nor can one turn to ancient Hebrew sources such as
the biblical quotations in the Talmud, for these Hebrew sources more or
less reflect MT. Biblical quotations in the Apocrypha are of equally
limited value, as most of these have been preserved in translation.
Thus there was no outside source which could support the retroversions
from the LXX, even though some evidence was available. For example,
the medieval Hebrew text of Ben Sira, known since the end of the last
century, provides some evidence that helps in the reconstruction of the
parent text of the Greek translation of that book. Furthermore, for the
reconstruction of the Vorlage of the LXX, important evidence may be
derived from the SP and the Hebrew context in MT.
The SP, known in the west since 1616, frequently agrees with the
LXX against MT, in pluses, minuses, and differences. One is often
tempted to reconstruct the Vorlage of the LXX to a form which is iden-
tical to the SP, and in such cases the latter thus serves as a support for
the former.’ For example:
Exod 1:22. MT QaD>wn aN) maledsrn ep Cee
SP (pawn Aen) oY? 319 Jan 2D
LXX tav dpoev 8 éav Te yb} Tots ‘EBpators (= TY, TI)
Exod 2:21 MT mw? Wl AT AX 7
SP TWN? AWN? IAD NX JN"
LXX kal €€€80T0 LeTIdwpav Thy Bvyatépa avtot
Mwuvoy yuvaika = S VMSS
Num 3:12. MT ON Ww? Wn AN ws 7WD2 75
SP YP OFT Dew" 399 ONT AD 732 75
LXX +AdTpa abtuv EoovtTa (cf. vv. 46 FF.)
Num 14:12) MT (1D?) TR WYN)
ol TaN PD ANT INR WYN)
LXX kal Totfow oé Kal Tov olkov Tot Tatpds Gou
Num 14:18 MT JOM IN OPDX TAR “7
3 This type of support is important, although Kyung-Rae Kim has shown that the LXX
and the SP were not as close to each other as was thought previously: Studies in the
Relationship between the Samaritan Pentateuch and the Septuagint, Unpubl. diss., Hebrew
University, Jerusalem 1994.
CONTRIBUTION OF THE QUMRAN SCROLLS 287
Sle NXT TON AN OD JAN ‘7 = Exod 34:6
LXX +kal ddnOLvds
In the past, however, these unique agreements between the LXX and the
SP were not invoked in support of the procedure of retroverting variants
from the LXX. They may have been overlooked or remained unmen-
tioned because most of the pluses consist of harmonizing elements which
are by definition secondary. Scepticism has often caused scholars to
disregard these unique agreements between the LXX and the SP in the
1900 instances listed by Castellus.4 Adherence to stereotyped views of
recensions and text-types of the biblical text has led scholars to
disregard these agreements, claiming that either the LXX was
translated from the SP, or that the SP was revised according to the
LXX, or conversely that the LXX was revised according to the SP.®
A second source of external support for the retroversions from the
LXX pertains to MT itself. Often a word in the context or in a parallel
section or book provides support for a retroversion. Thus the LXX of
Samuel and Kings occasionally reflects Hebrew variants identical with
parallel elements in Chronicles. At the same time, these parallel data
are often problematical. Inner-Septuagintal influences were at work,
and hence synoptic agreements between the LXX of Samuel-Kings and
the MT of Chronicles could be secondary. This pertains also to agree-
ments between the LXX of Chronicles and the MT of Samuel-Kings:
2Sam5:9 MI 3720 719 ae ae Re
LXX kal €xA)8y att h dds Aaud. kal Mkod6unocev
THY TOALY KDKAQ)
1 Chr 11:7-8 MT aaron Py7 jan Prt ey wap 1D by
2S5am5:21 MI (WIN) TIT ONW) O72XY NX OW IY
LXX kal Katadtuttadvouoty éxet Tous Peous auTuV
1 Chr 14:12 MT 9 wrawoe... o7772N DR OW 1aTY"
1Chr10:6 MT yn nw>w Dew no
LXX + €v TH MPEpa Exelyy
Cf. 1 Sam 31:6 Nw or2... ma nwa dwew na
Similar support can be drawn from the context in which the deviation
of the LXX occurs. E.g., the Greek plus in the LXX of Judg 16:13 can
4 The figures are based on the data in the sixth volume of the London Polyglot (1657),
part IV, 19-34. For a more recent listing, see Z. Metal, The Samaritan Version of the Pentateuch
in Jewish Sources (Hebr.; Tel-Aviv 1979).
Thus L. de Dieu, Seldenus, Hottingerus, and Hassencampius; for a detailed description
of their views and bibliographical references, see Gesenius, Pent. sam.
6 Thus Gesenius and Usserius; see Gesenius, Pent. sam., 13.
288 CHAPTER TWENTY
easily be retranslated on the basis of words occurring in the context.’ At
the same time, the context can often be misleading, for a Greek change
or plus, which is phrased like other elements in the context, could
reflect the translator’s manipulation.
Not until the discovery and analysis of the Qumran scrolls did any,
more direct, evidence for retroversion become available. Accordingly,
we now turn to the support from Qumran for the retroversion of variants
from the LXX, first some less known examples. In the following
example, an analysis of the Greek data allows for several possibilities,
while the Qumran data tilt the evidence in a certain direction.
One of the central formulae of Deuteronomy is ‘the land which you
(singular/plural) come to inherit.’ The two verbs used for ‘to come’ are
N12 and 72y, the latter one referring to the transgressing of the Jordan
prior to the coming into the land.
1. anw7> qow (o°X2 ONX) 8D ANN WR TATNA/ Pp INA (4:5; 7:1; 11:10, 29;
23:21; 28:21, 63; 30:16)
2. anw 2 maw (aay onX) WY ANN Wk TATNA/PpINA (4:14; 6:1; 11:8, 11)
The latter is a shortened formula of (Q°02¥ OX) 729 TNX WR WNTNT/ PINT
mnwt? maw (8127) 7777 MEX, found in 30:18 as well as in 4:26, 31:13 and
32:47 (in the latter three verses without 812).
The Greek translator of Deuteronomy distinguished between x11 and
jay, represented by elotropevopat and d:aBalvw respectively. There are,
however, four exceptions. In 4:14, 6:1 and 11:11 elotopetopat is used for
lav and in 11:29 S:aBalvw is used for x12. In view of the different
Hebrew formulae, these four exceptions could reflect inner-Greek
harmonizations, but since the translation of Deuteronomy is relatively
consistent, it is more likely that they represent Hebrew variations
between 811 and 719. This view is now supported by Qumran evidence:
6:1 o-nayv |] 4QPhyl B, M o-x2; 8OPhy! o°x3 = LXX elotopeteote.
Although independent harmonizing changes in the LXX, in 4QPhy]
B, M, and 8QPhy]i are not impossible, the assumption of actual variants
is more likely.
Note also the following two examples from Deuteronomy:
Deut 5:15 MT = nawa or ne mvy?
LXX +kal ayidCeiv abtihv
4QPhy] B (text of Deuteronomy) + v7?) = Exod 20:11
The added word in 4QPhy]l B supports the retroversion of the LXX as W777".
Deut 6:2 MT V8) "DIX WR
LXX + o7pLeEpoV
7 See TCU, 64-66.
CONTRIBUTION OF THE QUMRAN SCROLLS 289
SP, 4QPhyl J,M, 8OPhyl + ora
Cf. also the LXX in the following two verses:
Deut 12:11 MT = o9n8 789 7338 WR 79
LXX + o7LEpov
Deut 12:14 MT 18 7DIN WR
LXX + a7LEpov
In these three instances the LXX may have added onpepov on the basis of similar
phrases (4:40; 6:6; 7:11; 8:1, 11; 10:13; 11:8; 13:19, etc.), one of which is in the
context (6:6), but the readings in the phylacteries make it more likely that the LXX
actually reflects a Hebrew variant.
Unique agreements between the LXX and the Qumran scrolls, like
those mentioned above, abound in all books of the Bible. The reason
that a relatively small amount of such evidence is known is that but a
limited number of texts have been preserved in the Judean Desert. __
These agreements with Qumran scrolls increase our confidence in the
procedure of retroverting. Before 1947, retroversions from the LXX had
been attempted by generations of scholars, and therefore it causes no
surprise that variants reconstructed from the LXX by such scholars as
Thenius, Wellhausen, and Driver® have now actually been found in the
Hebrew Qumran scrolls, e.g. with regard to 4QSam?:
1Sam 1:23. MT 7727 NN 77 (OP" 7X) (= TV)
4QSam4 75 8800 ALT OP? FN]
LXX 10 €E€e ABV Ex Tot oTdyaTds Gou
Thenius reconstructed the LXX as 7°59 X¥777 NX.
1Sam 1:24 MT mw ow O92 = qwewn7|a (= T V)
LXX évudoyw TpretiCovTi = w2wN TDI = wPwWNID] = 5S
4QSam?2 wown aAPal
Thenius, Wellhausen, and Driver reconstructed the LXX as w?wn 7)2.
1 Sam 20:30 MIT nv MY) 72
... on of a perverse, rebellious woman (NRSV,
NJPST)
4QSam? n3v07 nw) 72
= LXX ulé kopaclwv avTopodotvTwy
... son of deserting maidens
Driver reconstructed the LXX as n171n(7) ny) 72.
Deut 32:8 MT IN TW? "ya (p07?) (= SP TOIFN S V)
(according to the number) of the sons of Israel . .
8 O. Thenius, Die Biicher Samuels (Leipzig 1842); Wellhausen, Samuel; Driver, Samuel.
290 CHAPTER TWENTY
4QDeud o77>x 713
(according to the number) of the sons of God
LXX848 106c ... uLav beod
LXXMost MSS dyyédwv Oeod = Aquila
For generations the readings of MT and LXX have been the topic of
much discussion.’ It is now evident that the LXX does not reflect
exegesis by the translators, but a Hebrew variant such as in 4QDeut.
See Tov, TCHB, 269.
Another intriguing group of examples pertains to small additions
and changes found both in a Qumran scroll and in the LXX, as
illustrated from 4QNum?. Until recently the text-critical value of the
LXX of Numbers was unclear. E.g.,
Num 12:6 MI NI WVDW TON)
LXX kal eltev mpd a’tovs dkovoate THY Adywv [LOU
It is difficult to determine whether tpds abtods reflects an added o77>x or the
translator’s exegesis, but the existence of this plus in 4QNum) strengthens the
former assumption. Similarly:
Num 16:5 MT (POR DP? 12) Ana (qwE nx)
LXX kal ols é&eAéEaTo Eautd TpoonydayEeTo Tpds
EQUTOV
4QNum? anja
Num 19:3. MT anX Onn
LXX Kal 8wWoeELs abT}V
4QNum? ane annn
Num 22:9 MT Wax") O92 OX OPN IM
LXX kal Adev 6 Beds Tpds Badaayp Kal eltev abTa
40ONum? + rox
Num 22:10 MT ON TW
LXX atéotetAev attous pds pe A€yuwv
4QNum? rx? ~>x|
Although the latter group leaves room for some doubt, hundreds of
examples enhance the credibility of the LXX as a text-critical tool in
biblical studies.19 They show that the intuition of generations of
scholars who ventured to reconstruct readings from the LXX, was
justified. In spite of known trends of exegesis in the translation, of inner-
translational corruptions and of our own ability to get back to the
9 See M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford 1985) 69.
0 This point has been stressed, among others, by Orlinsky, “Qumran” (above, n. 1);
Klein, Textual Criticism, 11-26 (esp. p. 13); and D. Barthélemy, “l’Enchevétrement de
l'histoire textuelle et de l'histoire littéraire dans les relations entre la Septante et le Texte
Massorétique,” in: Pietersma—Cox, De Septuaginta, 21-40, esp. 32-33.
CONTRIBUTION OF THE QUMRAN SCROLLS 291
Hebrew text underlying the translation, much of what has been done so
far in the area of retroverting the Vorlage of the LXX is now supported
by the Qumran finds, even though it should be borne in mind that each
book must be evaluated separately. 4QSam* has strengthened our
general confidence in the LXX of that book, and 4QJer>-4 support the
retroversion of the shorter text of the LXX of Jeremiah. At the same
time, not all agreements between the LXX and the scrolls against MT
are relevant to the discussion. Many a concurrence between 1QIsa@ and
the LXX may be coincidental and this may also be true for some of the
aforementioned harmonizations.
2. The recognition of a close relation between the LXX and specific
Qumran scrolls
Since many of the books of the LXX agree occasionally or frequently
with readings in Qumran scrolls, scholars have expressed opinions
about a specifically close relation between the LXX and some scrolls. As
a result, the term ‘Septuagintal scroll’ has made its entrance into the
scholarly literature. However, the establishing of such a close link is
beset with problems which relate not only to the facts themselves such
as the actual reading of the scroll and the meaning and reconstruction of
the LXX, but also to more general issues, such as the logic behind
statistical analysis and one’s overall text-critical Weltanschauung.
The following issues should be addressed beyond establishing the
reading of the scroll and the meaning of the words in the LXX.
1. It is often difficult to know whether a reading of the LXX which
differs from MT should be reconstructed as a deviating Hebrew reading
or should be regarded as the translator’s exegesis. In the latter case the
item should be disregarded. Exegesis which is common to the LXX and a
particular scroll is of interest, especially when occurring frequently, but
does not pertain to textual data.
2. It is often difficult to assess the extent of the agreement between
the LXX and a particular scroll. Does one count the items of agreement
separately and if so, how is this counting to be done? Usually, one counts
each agreement separately, including extensive textual phenomena
such as a long plus, minus, or difference. However, such items lose their
importance in a statistical analysis when they are included with items
of lesser magnitude. Agreements should therefore be subdivided into
those more and less significant.
3. The analysis centers on readings in which the LXX and a
particular scroll agree against MT. Within the web of the relations
between the textual witnesses there is something unusual in this
292 CHAPTER TWENTY
method of reasoning, to which we will soon return. But there is one
question which should be mentioned immediately: should we confine
our attention to exclusive agreements between the LXX and a scroll, or
should we include cases in which the LXX is joined by another ancient
version, such as S or T? For the sake of statistics such instances should
be included. The question is not very pressing, however, since most
instances pertain to exclusive agreements between the LXX and a scroll.
4. In the past, much stress was laid upon the counting of extra-
Masoretic agreements between the LXX and a specific scroll, while
disagreements were usually disregarded. The question arises as to
whether such analyses actually misrepresent the situation, especially
when there is an impressive number of disagreements between the two
sources. It seems that these disagreements do not necessarily diminish
the importance of the agreements, if the agreements are indeed
significant. Thus, if texts a and b are closely related in such a way
that they derived one from the other or from a common ancestor, either
a or b or both may have developed considerably since the stage at
which they were linked. Such subsequent development, now visible in
disagreements between a and b, should not necessarily undermine the
degree of affinity recognized between the two texts.
5. In determining the special relationship between the LXX and a
scroll, the textual character of the pericope or the book in question has
to be taken into consideration. If there is little textual variation in a
given unit, as in the case of the LXX and MT of Isaiah, the relation
between these two sources on the one hand and a Qumran scroll on the
other is bound to be very similar. Thus all the Isaiah scrolls from cave 4
(see DJD XV) agree with the MT and LXX almost equally, and
therefore their closeness to either the MT or LXX cannot be assessed.
6. As a rule, the determining of the relation between the LXX and
the scrolls does not take into consideration the originality of readings,
especially since such a question has very few objective aspects.
However, two exceptions should be made, relating to common secondary
and common original readings. With regard to the former, if the LXX
and a scroll agree in a presumed common secondary reading (often an
error), such an agreement may point to a very close connection between
the two. Such readings have been called ‘Leitfehler’ or ‘indicative
errors’ by P. Maas.!! However, in view of the fragmentary state of
preservation of the evidence it is hard to evaluate these Leitfehler.
The assumption of a close relation is possible, but one should realize
Il p, Maas, Textual Criticism, translated by B. Flower (Oxford 1958) 42 = Textkritik, in: A.
Gercke and E. Norden, Einleitung in die Altertumswissenschaft I, VII (3d ed.; Leipzig 1957).
CONTRIBUTION OF THE QUMRAN SCROLLS 293
that many other texts sharing these readings may have been lost. With
regard to shared original readings, if two texts share a reading which
probably is original, while the corrupted reading is found in another
source, the closeness reflected by the presumably original shared
reading is less significant, since it is natural for any two texts to share
original readings. This has become clear in particular with regard to
readings common to the LXX and 4QSam?. Thus the aforementioned
common reading of the LXX and 4QSamé? in 1 Sam 1:24 probably reflects
the uncorrupted text, while MT has been corrupted.!* This reading,
which must have been shared by additional texts which are now lost,
is thus less relevant to statistics, although only rarely can one
determine with certainty whether a reading is corrupt or original.
7. The coincidence of the textual transmission should be borne in
mind. Only some of the texts have been preserved, and accordingly
conclusions on the relation between the LXX, MT, and a scroll are
provisional, since in the hypothetical stemma of the manuscripts
several texts may have intervened between these three sources, while
others existed alongside these texts.
The agreements between the LXX and the scrolls have been collected
in the critical apparatuses of the editions of the Qumran fragments as
well as in separate monographs. Special attention has been given to a
few select scrolls. The relation between 1QIsa4 and the LXX has been
treated by J. Ziegler.!3 Likewise, much attention has been given to the
close affinities between the LXX and 4QSam? !4 and 4QJer>4.15
Very few Qumran scrolls are close to the LXX. In the assessment of
this closeness, the main criterion remains the subjective evaluation of
the number of agreements, subdivided into more significant and less
significant. It is hard to determine the lower limit of a list of
12 The common source of MT and the LXX (= 4QSam?) was almost identical: w>yn152.
According to the context, it is reasonable to assume that this word cluster originally
referred to a 15 (bull) in the singular (in the next verse the bull is referred to in the singular
[‘they slew the bull’]), i.e. ‘a-three-year old bull,’ and that when spaces between the words
as well as matres lectionis were added, the common source of the LXX and 4QSam® retained
this understanding, while MT was corrupted.
13 7, Ziegler, “Die Vorlage der Isaias-“Septuaginta (LXX) und die erste Isaias-Rolle von
Qumran (1QIs*),” JBL 78 (1959) 34-59.
4 See especially Cross, “New Qumran Fragment”; idem, “The Ammonite Oppression of
the Tribes of Gad and Reuben: Missing Verses from 1 Samuel 11 Found in 4QSamuel?,” in:
E. Tov (ed.), The Hebrew and Greek Texts of Samuel (Jerusalem 1980) 105-120 = H. Tadmor
and M. Weinfeld (eds.), History, Historiography and Interpretation Jerusalem 1983) 148-158;
Ulrich, Samuel; F.H. Polak, “Statistics and Textual Filiation: The Case of 4QSam2/LXX,” in:
Manchester Symposium, 215-276; E.D. Herbert, “4QSam? and Its Relationship to the LXX: An
Exploration in Stemmatological Analysis,” in: Taylor, IX Congress, 37-55.
ss) Janzen, Jeremiah; Tov, “Jeremiah”*; Min, Minuses and Pluses; Tov, “Jeremiah.”*
294 CHAPTER TWENTY
agreements establishing a close relationship. The most ideal case for
establishing a close relation between the LXX and a scroll is when the
scroll agrees with the LXX in readings that are characteristic of the
LXX.
The individual texts are discussed in a descending order of closeness
to the LXX, including a few texts that are not at all close to the LXX,
but nevertheless referred to in this context.
1. Two of the three fragments that have previously been labeled
4QJer®, and which are now named 4QJer® and 4QJer4,16 display a very
close relation with the Hebrew Vorlage of the LXX (unlike the other
scrolls of Jeremiah: 2QJer, 4QJer®‘). In fact, no other Qumran text is as
close to the LXX as these two fragments.
Characteristic of the LXX are the short name formulas, as opposed
to longer ones in MT, also found in 4QJer4 (e.¢., 43[50]:4, 5, 6).
Equally characteristic of the LXX are the long minuses and
differences in sequence, both of which are also present in 4QJer?
(Jeremiah 9-10). In this text, 10:6-8, 10 are lacking as in the LXX, and it
is probably impossible to reconstruct the order of the verses in 4QJer? in
any way other than that of the LXX, Le., 3, 4, 5a, 9, 5b, 11. The section
lacking in 4QJer® and the LXX (vv. 6-8, 10) has a uniform character: it
extols the Lord of Israel, while the remaining verses, extant in both the
MT and LXX, deride the idols of the heathen.
Both of these phenomena are amply described (see Tov,
“Jeremiah”*). There are some minor differences between the Jeremiah
scrolls and the LXX which show that the LXX was not translated from
the exact copy found at Qumran, but from a very similar one. Since the
agreements pertain to details which are characteristic of the LXX, it
stands to reason that the complete scrolls of 4QJer’4 would also have
agreed with the LXX in the chapters which have not been preserved.
2. The very fragmentary 4QDeut4 !” shares several important
agreements with the LXX. The most important agreement concerns four
additional stichs of the LXX in Deut 32:43, three of which are shared
with 4QDeut4, and none of which are found in MT. Several scholars
stressed the close relation between the LXX and this scroll.}8 Since
16 See DJD XV, 171-176, 203-205.
17 See P.W. Skehan and E. Ulrich, DJD XIV, 137-142.
18 The most recent and detailed statement is that of P.-M. Bogaert, ’Les trois rédactions
conservées et la forme originale de l’envoi du cantique de Moise (Dt 32,43),” in: N. Lohfink,
(ed.), Das Deuteronomium, Entstehung, Gestalt und Botschaft, BETL 68 (Leuven, 1985) 329-
340. For earlier discussions, see especially P.W. Skehan, “A Fragment” (n. 19); E.A. Artom,
“Sul testo di Deuteronomio XXXII,37-43,” Rivista degli studi orientali 32 (1957) 285-291; R.
Meyer, “Die Bedeutung von Deuteronomium 32,8f.43 (4Q) fur die Auslegung des
CONTRIBUTION OF THE QUMRAN SCROLLS 295
4QDeut49 ends with Deuteronomy 32 (and not with chapter 34), the
complete scroll probably was an excerpted text containing only the
poem of Deuteronomy 32. Our conclusion regarding the textual character
of 4QDeut4 thus probably pertains not to that scroll only, but it may be
extended to the text from which 4QDeut4 was copied.
3. The proximity of 4QSamé to the LXX and LXX"" has been stressed
frequently. Note especially the title of the study by F.M. Cross, “A
New Qumran Fragment Related to the Original Hebrew Underlying
the Septuagint,” BASOR 132 (1953) 15-26. The number of the agree-
ments between the LXX (including LXX/"°) and 4QSam? is impressive,
judged on the basis of the publication of the first two columns by Cross
and of scattered readings listed by Ulrich!9 and McCarter.*° The
statistics for these agreements are as follows:?!
Col. I (1 Sam 1:22-2:25) 4QSam4 = LXX® # MT 22 (possibly: 28)
4QSam?2 = MT #LXX® 4
4QSam4 # LXXB#MT 5
4QSam? #LXXB=MT 9
When the data for LXX®8 and LXX/" are separated, the relation
between the sources for this column is expressed as follows:
4QSamé@ = LXXB 22 (possibly: 28) 4AQSam? # LXX8 18
4QSam4 = LXxXbuc 17 (23) 4QSam? #4 LXXL“< 20
4QSam2 = MT 4 4QSam? # MT 41
In the second column of the scroll the statistics are as follows:
Col. II (2 Sam 3:23-5:14) 4QSam2 = LXX® #MT 13
4QSam4 = MT #LXX8 7
4QSam4 # LXXB# MT 4
4QSam4 # LXX8=MT 6
The relation between the sources is expressed as follows:
4QSam? = LXXB 13 40Sam4 #LXxXB 22
4QSam@ = LXxbLuc 13 AQSamé4 # LXXLuc 15
4AQSam2 = MT 10 4Q0Sam? # MT 25
Mosesliedes,” in: A. Kuschke (ed.), Verbannung und Heimkehr. Beitrage ... W. Rudolph ...
(Tiibingen 1961) 197-209.
19 Ulrich, Samuel.
20 P.K. McCarter, I-I] Samuel (AB; Garden City, NY 1980, 1984).
21 Listed in Tov, “Methodology.”*
296 CHAPTER TWENTY
The closeness between 4QSamé4 and the LXX is thus impressive at the
statistical level, even more so in the figures given by Polak and
Herbert, who refer to all known readings of the scroll (see n. 14 ).
From this substantial number of agreements one has to deduct
readings common to the LXX and the scroll which presumably reflect
the original text, against a corrupted form in MT. Long before the
discovery of the scrolls, such scholars as Thenius, Biicher Samuels,
Wellhausen, Samuel, and Driver, Samuel had recognized the often
faulty character of MT. Accordingly, the fact that the joint readings of
the LXX and 4QSam/4 often contain original elements does not prove a
particularly close relation between these two sources. Many other texts
may have contained those readings, while MT, being the exception,
contained erroneous details. Since many ancient scrolis of Samuel have
been lost, comparison between the now preserved witnesses, LXX, MT,
and 4QSam, presents data which can easily be misjudged because of
the optical illusion presented by the fragmentarily preserved
evidence.?2
The LXX and 4QSamé also differ significantly and both contain
exclusive readings, such as the long plus in 1 Samuel 11 in 4QSamé@ (see
TCHB, 342-344) and its divergent version of the Song of Hannah (1 Sam
2:1-10)—see Tov, “Song of Hannah.”* However, in the putative
stemma of the manuscripts there is room for such exclusive readings if
they were created after the point at which the texts separated from
each other.
The LXX and 4QSam# contain a few readings which P. Maas would
call ‘indicative,’ viz., leading common errors. This pertains to the
extensive doublet in 1 Sam 2:23-24 and to the erroneous mention of
Mephibosheth’s name in 2 Sam 4:1, 2 ,12 as opposed to Ishboshet in MT
in v. 12 and the absence of a name in wv. 1, 2. These significant common
errors suffice to establish a close connection between the LXX and the
scroll, but this connection is not as close as in the case of the two Qumran
texts mentioned above. Beyond the aforementioned original readings
22 Iisa very subjective and difficult matter to earmark a certain variant as original, and
the reading of MT as an error, but there are quite a few instances of such evaluations in
the book under consideration, which single out 4QSam® as a very special text. Thus, the
shorter text of MT in 1 Sam 1:24 is usually recognized as having been omitted by way of
homoioteleuton as compared with the longer and original text of 4QSam® and the LXX.
The aforementioned reading in 1 Sam 1:23 is likewise considered to be a mistake in MT. In 1
Sam 2:22 the LXX and 4QSam@ also contain the original short text while the expanded text
of MT (v. 22b) has been recognized by most scholars as a theological gloss. A certain number
of the common readings of the LXX and 4QSam® have thus to be deducted from the list as
less relevant details.
CONTRIBUTION OF THE QUMRAN SCROLLS 297
and common errors, the agreements between the LXX and 4QSam?
include single details which are not characteristic in any way. The
relative location in the putative stemma of 4QSam@ and the Vorlage of
the LXX cannot be further determined.
4. 4QNum) (see DJD XII, pp. 205-267) contains an impressive list of
agreements with the LXX. There is no common denominator for these
agreements, but many of them are small harmonizing pluses based on
the context. Several of these extra-Masoretic agreements between the
LXX and the scroll are shared with the SP, and actually, the scroll
displays a greater similarity with the latter version than with the
LXX. Among other things, it shares with the SP the major harmonizing
pluses based on Deuteronomy (Num 20:13; 21:11, 12, 20; 27:23).
5. 4QExod? (previously named 4QEx“) contains a substantial number
of agreements with the LXX (analyzed by F.M. Cross in DJD XII, 79-95).
6. 4QLev¢ (Leviticus 14-17) agrees twice with the LXX in long pluses
and three times in small details, while disagreeing in two minor
details. Although the text is not extensive, its affiliation is primarily
with the LXX, secondly with the SP, and only thirdly with MT.
The major plus in 4QLev¢ in Lev 17:4 is shared with SP and the LXX,
and therefore the connection between the scroll and the LXX is less
exclusive than in the other instances. This plus is probably secondary.”3
Three scrolls show an impressive degree of agreement with the LXX
in small details at the statistical level, but this agreement is probably
misleading:
7. 4QDeut* contains 12 exclusive agreements with the LXX as well as
19 instances of disagreement (see DJD XV).
8. 4QDeut® contains 9 exclusive agreements with the LXX as well as
9 instances of disagreement. Of particular interest are agreements
between the LXX and 4QDeut) in Moses’s blessing in Deut 33:8-11,
which are shared with the quotation in 4QTest (4Q175) against MT.74
9. 4QDeut) agrees 4 times with the LXX against MT, while
disagreeing 6 times with that source.
The following texts have been mentioned as being close to the LXX,
but the evidence is not convincing.
23 The text expands v. 4 based on v. 3, and therefore does not add substantial details to
the implication of the law. See earlier commentators: B. Baentsch (HKAT; Gottingen 1909)
389; E. Elliger, Leviticus (HAT; Tubingen 1966) 219. According to E. Eshel, “4QLev¢: A
Possible Source for the Temple Scroll and Migsat Ma‘ase Ha-Torah,” DSD 2 (1995) 1-13, this
text was probably also known to the authors of the Temple Scroll and 4QMMT.
For a detailed analysis, see pp. 34-35 of the original article as well as S. Beyerle,
“Evidence of a Polymorphic Text—Towards the Text-history of Deuteronomy 33,” DSD 5
(1998) 215-232. Beyerle terms this text non-aligned (p. 232).
298 CHAPTER TWENTY
10. Milik’s contention*? that 5QDeut (chapters 7 and 8) has been
revised four times according to a Hebrew text close to the Vorlage of the
LXX would have been of special interest had the evidence been more
conclusive. Indeed, two of the corrections agree with the LXX against
MT (the addition of Wx) amx 7 in 7:15 and that of 02 in 8:12). The third
correction (8:19) is based on a reading which at best is dubious, while
the fourth instance is probably irrelevant (9:2). At the same time, there
are eight instances of disagreement between the LXX and 5QDeut and
two agreements in minutiae. The sum of this evidence does not favor the
conclusion that this text has been corrected towards a Hebrew source
close to the LXX. In fact, no Qumran manuscript has as yet been
identified in which corrections clearly tend towards either the LXX or
MT.6
11. 4QSam<¢ is equally close to MT and LXX/"° in 2 Samuel 14-15,
which in that section probably reflects the OG.’ It is less close to the
main tradition of the LXX which in these chapters contains kaige-Th.
At the same time, lack of evidence warns us not to draw any special
conclusion concerning a specially close relation between the LXX and
the Lucianic or Old Greek text of Samuel.
12. The closeness between 1QIsa@ and the LXX was discussed soon
after the discovery of the scroll,28 but the arguments adduced by
Ziegler (see n. 13) show that there was much exaggeration in these
early observations. Most of the agreements between the two sources are
in minutiae, and as Ziegler realized, they may be coincidental.??
25 DID Itl,169-171.
See my article “The Textual Base of the Corrections in the Biblical Texts Found in
Qumran,” in: D. Dimant and U. Rappaport (eds.), The Dead Sea Scrolls—Forty Years of
Research (Leiden/Jerusalem 1992) 299-314.
27 See the analysis in Tov, “Methodology,”* esp. 58-61. Ulrich, on the other hand,
particularly stresses the links with the Lucianic tradition: “4QSam‘°: A Fragmentary
Manuscript of 2 Samuel 14-15 from the Scribe of the Serek Hay-yahad (1QS),” BASOR 235
(1979) 1-25.
Upon the publication of this scroll, scholars were quick to remark on its agreements
with the LXX. For references, see the discussion by Orlinsky, “Qumran” (n. 1 above) and
see also Orlinsky’s own analysis. The agreements between the Isaiah scroll and the LXX
have been listed and analyzed by many scholars: M. Burrows, BASOR 111 (1948) 20-24; J.T.
Milik, Bib 31 (1950) 86; D. Barthélemy, RB 57 (1950) 539-543; G.R. Driver, ]ThSt NS 2 (1951)
25ff.; D.M. Beegle, BASOR 123 (1951) 26-30; M.H. Goshen-Gottstein, Bib 35 (1954) 51-71; O.
Léfgren, Donum Natalicium H.S. Nyberg Oblatum (Uppsala, 1954) 180ff.; P. Wernberg-Maller,
]SSt 3 (1958) 254-264; Orlinsky, “Textual Criticism”; Kutscher, Isaiah Scroll.
29 That is, the small contextual changes such as in number, pronouns, particles and
verbal forms, which the two sources sometimes have in common could have developed
independently.
CONTRIBUTION OF THE QUMRAN SCROLLS 299
13. 2QDeut® has been described as follows by Baillet: “Le texte se
rapproche de la LXX et de la Vulgate.’39 However, this fragment, of
which a mere twelve words have been preserved, in whole or in part,
shows no close relation to either the LXX or Vulgate.?!
3. The Qumran scrolls and the history of the biblical text
Soon after the discovery of the first Qumran scrolls scholars referred to
the close relation between some Qumran scrolls and the LXX, although
at the time this issue was not examined closely. The recognition of this
proximity should be viewed within the wider picture of an analysis of
the relations between textual witnesses. It was natural then, and
remains so today, for many scholars to describe the history of the
textual witnesses of the Bible in terms of three recensions, families, or
revisions, at the center of which stand the MT, LXX, and SP (see TCHB,
chapter 3A). It was also natural for scholars to try to ascribe the newly-
found texts to one of the three groups/recensions, since no entity beyond
this tripartite division had been envisaged. Thus most of the Qumran
texts were ascribed to the so-called recension of MT, some to that of the
SP, and yet others to the recension of the LXX, also named the
‘Egyptian family.’ In our view, however (see TCHB, chapter 3A), the
three textual entities are not recensions, but rather texts, and more than
three such texts are known. But against the background of the scholarly
consensus on the status of the textual witnesses, it is easy to understand
how and why certain scrolls were ascribed to the ‘recension’ of the LXX,
and subsequently described as ‘Septuagintal.’
If a Qumran text could not be ascribed to the recension of MT,
according to the scholarly consensus it almost had to be ascribed to that
of the SP or the LXX. There was no room for the assumption of a
different status of the scroll. In light of this it is understandable why
at first scholars thought that 1QIsa? was close to the LXX—after all,
its text is not particularly close to MT, and there seemed to be only one
alternative, viz., to assume a close relation with the LXX. Milik’s
contention that 5QDeut was revised according to the LXX reflects a
similarly limited textual outlook, and the same pertains to Baillet’s
remarks on 2QDeut‘. The same view underlies Milik’s description of
5OKings: ‘Le peu de texte conservé n’est pas significatif du point de vue
recensionel: le TM et la LXX y sont a peu pres identiques’ (DJD II, 172).
Also some of subsequent publications display a similar outlook.
30 DID IH, 61.
1 1¢ anything, this text, written in the “Qumran orthography,’ agrees more with MT
against the LXX than vice versa.
300 CHAPTER TWENTY
The issue at stake is one of statistics and textual outlook. A number of
agreements between a scroll and the LXX does not make that text close
to the LXX or ‘Septuagintal,’ so to speak, even if the list is impressive
(such as in items 7-9 above). The LXX is just a text and not a recension. A
large number of agreements with the LXX only shows that the two texts
are closer to each other in the supposed stemma of the biblical texts
than to the other known texts. Even if we do not succumb to stemmatic
considerations for the biblical texts, there is nothing wrong in doing so.
However, with the enormous gaps in our knowledge we will never be
able to assess the real relation between the texts.
Many of the calculations of the closeness between the LXX and a
scroll are based on the accumulative evidence of many readings,
sometimes important, but often not. The tacit assumption behind this
thinking is, as mentioned, that there were merely two or three
recensions and that simple statistics can show us how close the Qumran
text was to one of the three or two recensions of the biblical text.
However, these texts were no recensions, but simply texts, and their
number was at one time much larger than two or three. Moreover,
probably only a very small number of these texts is known to us. As a
result, any speculation on the basis of these very fragmentary data may
be misleading if based on mere statistics.
Since only a few Qumran texts are close to the LXX, a term like
‘Septuagintal scroll’ should be avoided. That term is based on the
wrong assumption that the Septuagint reflects an archetypal recension
of the biblical text.
At Qumran only a very small number of texts was found that were
closely related to the Vorlage of the LXX (less than 5 percent of the
biblical texts). The Hebrew scrolls from which the LXX was translated
in Egypt have not been found at Qumran. Since many, if not most, of the
biblical texts of the third and second centuries BCE were unique, they
should be sought only in Egypt itself, even though they were originally
imported to Egypt from Palestine.
There is insufficient evidence for speculating on the relation between
the individual Qumran texts which are close to the LXX. They should
not be considered a group, they do not form a textual family like the
MT-group or the proto-Samaritan texts, and they have not been
produced by a scribal school. They represent individual copies that in
the putative stemma of the biblical texts happened to be close to the
Hebrew text from which the LXX was translated.
CHAPTER TWENTY-ONE
INTERCHANGES OF CONSONANTS BETWEEN THE MASORETIC
TEXT AND THE VORLAGE OF THE SEPTUAGINT
1. Background
The biblical text consists of many details, each of which is usually
called a ‘reading.’ When sources differ from one another, these readings
are ‘different’ or ‘variant’ readings, or simply ‘variants.’ There are
thousands of such variants, subdivided into pluses, minuses, trans-
positions, and differences. The last category forms the topic of the
present paper, and since so many differences exist among the textual
witnesses, MT (codex L) is being used as the basis for all comparisons.
This study focuses on the differences between MT and variants
reconstructed from the LXX. These reconstructed variants are as
important as Hebrew variants, as long as they are based on reliable
reconstructions.
The thousands of details in the LXX whose reconstructed parent text
differed from MT are called ‘reconstructed variants,’ but the terminol-
ogy may be confusing.! This term may give the wrong impression that
every reliably retroverted variant was once found in the Hebrew scroll
from which the translation was made. However, even if a variant
seems to be reliably retroverted from Greek, that reading may never
have existed anywhere but in the translator’s mind. This applies in
particular to retroverted variants which are mistakes, for as a rule one
cannot know whether the mistake was made by the translator who
misread his Vorlage, or by the scribe of the Hebrew source. E.¢.:
Jon 1:9 MT (87778 QyowiT PX “7N)) PDR DY (OPN WN)
SotiAos Kuplou Ey elt
For MT -729, the LXX has 5000s kuplov which may be retroverted as
mit Tay. In this instance the LXX probably reflects an abbreviation of
the tetragrammaton as yod, as well as a dalet—res interchange. This
lee. Barr, Comparative Philology, 238-239; Tov, TCU, 88-89.
302 CHAPTER TWENTY-ONE
retroversion probably yields an inferior reading, but this issue is not
important in the present context. The central point in the argument is
that there once existed a tangible or abstract reading 72, even though
we do not know at which level that reading was created. One
possibility is that the translator found in front of him a reading ‘* 729
in which the yod indicated (or was thought to indicate) an abbreviated
tetragrammaton. However, possibly such a reading “* 729 never existed
in reality, but only in the mind of the translator who mistakenly took
the letters *72¥ as “* 72y. Likewise, any misreading of 07x (Aram =
Syria) as O(1)78 (Edom), or vice versa, could have occurred in the scroll
from which the translation was made, but by the same token it could
have existed only in the translator’s mind.
There are hardly any criteria for distinguishing between retroverted
variants that existed in writing and those that existed only in the
translator’s mind. This point should be stressed, because not all scholars
use the ambiguous term ‘variant’ in the same way. Some use the term
‘variant’ or Vorlage with the implication that each retroverted
reading actually existed in writing. However, even the most reliable
retroversions of variants refer to readings that may not have existed in
writing. Unfortunately, due to lack of suitable controls, retroverted
variants that existed only in the mind of a translator are also called
variants. Accordingly, when referring below to variants reflected in the
LXX, I refer to variants which are reliably retroverted, but may not
have existed in reality.
The present article focuses on interchanges of single letters or
combinations of two letters, mainly because of their similar shape. This
focus thus excludes interchanges of letters or words because of other
reasons, such as context, language, exegesis, and theology. The
distinction is often difficult. For example, is the interchange between
v1 and 1>uP" in a given context linguistic or did it derive from a scribal
mishap? The present article does not deal with linguistic interchanges,
but since it is often hard to distinguish, some of these are included as
well. Interchanges of synonymous pairs (e.g., ?XTW? "12 / 2NTw? M3) are
not included. But the following instance is included: In Jer 2:18 the LXxX
reads I‘nwv, which in Hebrew would point to 7)n1 instead of T1nw in the
MT’s 7inw 79 minw? (o-RD 4779 72 AN Anyi). The LXX has Tod tretv tdwp
[nwv. Does Pqwv imply interchanges of sin and gimmel and of res and
nun on the scribal level, or a substitution of two geographic entities? A
decision is difficult, but the scribal interchange sin/gimel is at least
possible, and that or res /nun is very likely.
INTERCHANGES OF CONSONANTS 303
2. Interchanges
This study examines differences in consonants (letters) between MT and
the reconstructed Vorlage of the LXX, that is, when one letter is
replaced with another one or two, presumably because of their external
similarity. Added and omitted letters such as iyvnw/inw or ynw?/ynw are
excluded. The following aspects are examined:
a. Which letters interchange on the scribal level?
b. What is the frequency of a given interchange in an individual
biblical book?
c. Can we point to certain stages in the development of the paleo-
Hebrew and/or square scripts reflected by the interchanges in the
LXX?
d. Do certain books reflect more interchanges than others?
3. Proper nouns
The most fertile ground for the study of interchangeable letters both in
Hebrew sources and between the Hebrew and Greek sources is the realm
of proper nouns, since in this area very few content considerations are
involved.
For example, in the parallel lists of David’s mighty men in 2 Samuel
23 and 1 Chronicles 11, the following pairs of names refer to the same
persons, even though their name and provenance vary slightly.
2 Sam 23:25-26 D207 PIN... "TANT Aw
Chr 1327 “DT PAM TIAA mw
The interchanged letters mentioned are known from many other places.
ow and nww imply an interchange of he and taw, with an added waw
as mater lectionis in Chronicles. Furthermore, the provenance of this
person is 1n according to Samuel, but 177 according to Chronicles. Here
we encounter interchanges between he and het and between dalet and
res, again with an added waw in Chronicles. Also the provenance of
the second person, yon, differs in the two sources. In Samuel he is
referred to as "D7 and in Chronicles as 711957, reflecting an interchange
between fet and a waw-nun ligature.
2 Sai 23:27 "Nwnn 7 nny WylaNX
1 Chr 11:28—29 "NWA D320 “NINA WYN
This example is more complicated. Was the second ‘mighty man’
named Mevunay as in Samuel or Sibkay as in Chronicles? At the level
304 CHAPTER TWENTY-ONE
of pronunciation the difference is considerable, but it is slight at the
level of the consonants.
Similar interchanges are found between the MT and LXX and they
add to our knowledge of what happened in ancient Hebrew manus-
cripts. The evaluation of the Greek form of these names is, however, not
easy since scribes who did not know Hebrew corrupted many a name.”
Thus there are many cases in which the exact Vorlage of the LXX can no
longer be reconstructed, such as:
1 Kgs 4:10 TDM PAN 7D) 1Dw 1?
Aovoayunvyxa kal Pnodapa
The translator took 1? as part of the name. As for the transliteration of
the name, he probably read one or more letters between the kaph and
the he. A case such as this is not included in the present statistics.
Josh 10:3 Orn»
®Lbwv
The Greek translator read a dalet for the res of MT, but it is unclear
whether the end of the word reflects a variant or an inner-Greek
variation. The dalet—res interchange is included in the statistics below.
Josh 15:23 yan) En)
Kal AOOPLWVALV
The two words have been combined to one; it is unclear how the first
three letters of the second word were read by the translator. This word
is excluded from the statistics.
A more advanced knowledge of the possible interchanges of Greek
letters eliminates incorrect reconstructions of Hebrew variants. Thus
uncial Jambda and delta are known to interchange in Greek
manuscripts, a situation which makes the assumption of a dalet/lamed
interchange unnecessary.
Ezek 27:16 nnyx71(1)
Aawwé BOL
Aap Cc
Pauwé 967
In order to understand what kind of interchange has occurred here, one
has to identify the original text of the LXX. Taking a clue from the
Hebrew text, both Papwn of LXX67 and Aayw6 of LXX© could reflect the
2 See F. Knobloch, Hebrew Sounds in Greek Script: Transcriptions and Related Phenomena in
the Septuagint, with Special Focus on Genesis, unpubl. diss., University of Pennsylvania
(Philadelphia 1995); Classified Bibliography, 41-42; see also the editions in the Gottingen
Septuagint series together with the accompanying studies.
INTERCHANGES OF CONSONANTS 305
original Greek reading, as they could reflect either the MT’s nnxn ora
variant nnx7, created by an interchange of dalet and res. Aauwd of BOL
could only have developed on the inner-Greek level from a reading
starting with a delta by the frequent delta/lambda interchange.
There-fore Aayw6 of C is taken as the original reading corrupted to
Aaywé in BOL, while Paywé is a correction toward MT in P. Chester
Beatty 967. If this reasoning is correct, the original reading of LXX
Aapwé thus reflects a dalet/res interchange (Cf. Ziegler, Beitrige, 61).
However, it often is difficult to determine whether a particular
interchange is inner-Greek or is between the Greek and Hebrew. A
particularly difficult case is the possible interchange at the ends of
words as either mem/nun in Hebrew or mu/nu in Greek. Ziegler suggests
that the Greek scribes were prone to interchanging these letters
without any connection to Hebrew sources, as in AxLKap/AXLKay,
Darkwy/Darwy, Duxew./Zvxev, EXvabav/EAvaGau. In all these cases some
Greek witnesses change a majority reading from mu to nu or vice versa.
However, at the same time a Hebrew mem/nun interchange is
evidenced in Hebrew sources, so that a number of these cases must be
ascribed to Hebrew variations. More detailed studies are needed on the
Greek sources, but many unusual mem/nun interchanges are evidenced as
well. From rabbinic Hebrew forms such as ])>v for 019v are known, as
well as 78 for 07% and 7°x for ox. Such forms are also known from the SP
and inscriptions from the Second Temple period. 1QIsa* has o>7” for 77
in 9:3 and 60:6, so that the frequent LXX transliteration Madcap (e.g.,
Gen 25:2, 4; 36:35) could be ascribed to a Hebrew rather than inner-
Greek interchange. The same applies to a long series of names such as
Teoep for 7w1, EyAwp for 71219 (Josh 12:12A), and Mappwv for an (Josh
L127).
A similar problem obtains with regard to the added nu in words
ending in MT with a vowel. One thus notices a frequent transliteration
of xn°n as Batuav (Jer 25:23 [32:9]; Isa 21:14), of Xaw as VaBav (Gen 25:3),
and of nwy>x as EXeaoav (Jer 29[36]:3b). Ziegler, ibid., 73 ascribes to the
view that this is an inner-Greek addition of a nu. However, the
alternative view cannot be discarded easily. jan next to Xn°n is known
from MT, and various biforms with a nun for nouns ending with vowels
are known from Hebrew sources. Thus }717° and ]77 are known next to
mn? (see Mur 18 in DJD II, 104). In all these cases we must consider the
possibility that the LXX reflected a Hebrew form ending with a nun.
3 Ibid., 66-67. Thus also Seeligmann, Isaiah, 65.
306 CHAPTER TWENTY-ONE
4. Common nouns
Beyond the aforementioned complications, retroversions from the LXX
based on transliterated Hebrew words are considered relatively easy
part of the act of retroversion. Much more problematic are the retrover-
sions of common nouns. The subjectivity of this procedure hardly needs
to be demonstrated here. One example suffices.
Jer 2:12 MT 38 TIN yw net oy ow ww
reconstruction TA TI ...
LXX €€éoty 6 otpaves ént TouTw kal
édpiEev én trAEtov o865pa
If the reconstruction of the Vorlage of 129n to 7270 is correct, it provides
information on interchanges between het and he and between waw and
he. If the retroversion is incorrect, by implication this information is
irrelevant.
5. The data
Information on interchanges of letters between the MT and LXX,
involving proper and common nouns is provided below, based on
subjective reconstructions (for proper names reasonable objectivity can be
obtained).
Table 1 records the data on these interchanges, book by book, giving
the total number of interchanged consonants in a given book, counting,
for example, 12y for 73y (such as 72y for 71y), but disregarding an
interchange ?X7w> 7}2 and Ox7w m3, since presumably this is not a
scribal phenomenon. Interchanges of person and form in the verb as well
as prepositions (e.g., ynwi/ynw) are disregarded, since these are too
uncertain. The data are culled from the CATSS database (see Tov,
“Computerized Data Base”’*). The major types of interchanges, dalet/
res, yod/waw, and mem/nun, are listed in Table 1. The last column
provides comparative statistics regarding the total number of assumed
interchanges compared with the overall number of words in any given
book.
The same data are repeated in Table 2, re-arranged according to the
relative frequency of the interchanges in the books.
INTERCHANGES OF CONSONANTS
Table 1
Interchange of consonants in the individual books
307
Total Number Number of
of Words in
Book 1/4 By 1/ _Interchanges Book? Percentage
Gen 22 | 27 106 20,613 0.51
Exod 5 2 - 20 16,713 0.12
Lev 2 1 1, 22 11950 0.18
Num a7 7 ame al 16,408 0.43
Deut 3 i] 4 35 14,294 0.24
Josh 29 i2 51 166 10,051 1.65
Judg A 6 1 - 2] 9,886 0.21
Judg B 3 1 =: 5 9,886 0.15
1Sam 29 6 2 83 13,264 0.63
2 Sam a2 10 - 100 11,040 0.91
1 Kgs 33 8 5 104 13,140 0.79
2 Kgs 12 12 14 88 12,284 O72
Isa 38 20 4 98 16,934 0.58
Jer 55 ZA 5. 190 21,836 0.87
Ezek 38 13 16 164 18,730 0.88
Hos 15 10 - 60 2,381 252
Joel 2 1 - 4 957 0.42
Amos 7 4 - 19 2,042 0.93
Obad 3 3 - 7 299 2.34
Jonah - - - - 688 0
Mic 12 3 - 21 1,396 1.50
Nah 2 1 - 12 558 2215
Hab 2 1 - 6 671 0.89
Total Number Number of
of Words in
Interchanges Book?
Book 7/5 a i/n Percentage
Zeph 1 1 - 3 767 0.39
Hag : - - 600 0
Zech 1 1 - 5 3,128 0.16
Mal S, - - 6 876 0.68
Ps 10 14 - 46 19,587 0.23
Job 8 2 - 19 8,351 0.23
Prov 29 16 - 115 6,915 1.66
Ruth - - - 4 1,296 Os
Song - - - 5 1,250 0.40
308 CHAPTER TWENTY-ONE
Qoh 3 : = LO 2,987 0.33
Lam 1 2 = 11 1,542 0.71
Esth> : - oe 3,045 0
Dan 6 2 D2 5,919 0.35
Dan-Th 2 1 ay, 5,919 0.12
Ezra 4 20 - 26 3,754 0.69
Neh 6 15 3 39 5,312 0.73
1 Chr 4] 57 25 178 10,746 1.66
2 Chr 4 5 s 17 13,315 0.13
a The data on the number of words for the individual books of the
Bible were provided by P. Cassuto of the Centre d’Analyse et de
Traitement Automatique de la Bible et des Traditions Ecrites in
Villeurbanne, France.
b Without Apocrypha
Table 2
Comparative analysis of the frequency of interchanges
Book Total Number of Number of
Interchanges Words in Book Percentage
Jonah - 688 0
Hag - 600 0
Esth? - 3,045 0
Dan-Th 7 5,919 0.12
Exod 20 16,713 0.12
2 Chr 17 135515 0.13
Judg B 15 9,886 0.15
Zech 5 3,128 0.16
Lev 22 11,950 0.18
Judg A 21 9.886 0:21
Job 19 8,351 0.23
Ps 46 19,587 0.23
Deut 35 14,294 0.24
Book Total Number of Number of Percentage
Interchanges Words in Book
Ruth 4 1,296 0.31
Qoh 10 2,987 0.33
Dan 2 5,919 0.35
Zeph 3 767 0.39
Song 5 13250 0.40
Joel 4 957 0.42
Num A 16,408 0.43
INTERCHANGES OF CONSONANTS 309
Gen 106 20,613 0:51
Isa 98 16,934 0.58
1 Sam 83 13,264 0.63
Mal 6 876 0.68
Ezra 26 3,754 0.69
Lam iB 1,542 0.71
2 Kgs 88 12,284 0.72
Neh 39 5,312 0.73
1 Kgs 104 13,140 0.79
Jer 190 21,836 0.87
Ezek 164 18,730 0.88
Hab 6 671 0.89
2 Sam 100 11,040 0.91
Amos 19 2,042 0.93
Mic 2.1 1,396 1.50
Josh 166 10,051 1.65
1Chr 178 10,746 1.66
Prov LES 6,915 1.66
Nah 12 558 215
Obad 7 299 2.34
Hos 60 2,381 2.52
a4 Without Apocrypha
Conclusions
1. As expected, there is no correlation between the frequency of
interchanged consonants as listed above and the frequency of pluses and
minuses in the LXX.
2. The statistics indicate differences in the frequency of interchanges
of consonants. The smaller the number of interchanged consonants, the
more evidence there is for a stable textual transmission and careful
translation. Most of the books have up to 0.5 percent interchanges when
compared with the number of their words, while a few have between
0.5 and 1.0 percent. Noteworthy are books with a greater number of
interchanges, recorded in the last seven lines of Table 2. Of special
interest is the high frequency of interchanged consonants in Hosea.
Many have claimed that the text of Hosea is corrupt, and the relative
frequency of consonantal interchanges there may help to confirm this
conclusion. The same applies to 1 Chronicles in contradistinction with 2
Chronicles, but the data may be misleading, for 1 Chronicles contains
more names than the other books of the Bible and much of the confusion
relates to the names. It is hard to evaluate the evidence of the other
310 CHAPTER TWENTY-ONE
books (Micah, Nahum, and Obadiah), since the absolute numbers are
small; yet, it should be noted that all belong to the Minor Prophets.
At the other end of the scale we note books which show evidence of
little scribal activity, as seen from the first nine lines of Table 2. This
group contains a few late books: Esther, 2 Chronicles, Daniel-Th, and
Haggai, as well as two books of the Torah.
3. The great majority of interchanges occur in every single book only
once or twice and occasionally three times. The only interchanges
which occur frequently in most books of the LXX are 1/7 and °/1. Some of
the numbers may be misleading as they include a large number of the
same names.
Thus of the 33 interchanges of 1/7 in 1 Kings, 14 refer to the change of
T1n 72 to 771n 72 (ulés Abep), and 6 are of 3737 to 177 (Adep). In 2 Kings
there are 4 similar instances of 377 72, as well as four cases of 2°7710/
Levvaxnpty and 6 of 773717" /] waxy. In 2 Samuel there are 9 instances of
1y77n rendered as ASpaacap. In Numbers, 14 of the 27 interchanges of
1/> refer to the interchange of fwii/jw7) and 71W71/7107).
4. Very few patterns in the direction of the interchange can be
detected. With one exception no cases were found in which there were,
for example, more interchanges from dalet in MT to res in the LXX than
from res in MT to dalet in the LXX. Usually the numbers are more or less
equal, for example, 27 instances of 1/7 between the MT and LXX of
Jeremiah as against 28 cases of 1/7. The one exception is in Ezra, which
contains 17 instances of an interchange °*/1 and only 3 of the reverse 1/°,
almost all in personal names.
5. In almost all books the only two interchanges occurring with any
frequency are 1/7 and */1 and in many books they occur with great
frequency. In 1-2 Samuel one-third of all interchanges are between 1/7,
and the same applies to 1 Kings and Isaiah. In Micah that number is
about one-half of the total instances. In Ezra, of the 26 interchanges, 4
are between 7/7 and 20 are of */1. Usually all other interchanges occur
with much less frequency.
In almost all books the interchange 1/7 is much more frequent than
that of °/). Exceptions are Psalms and 1-2 Chronicles with some
preponderance of the interchange */) over 7/7, and Ezra and Nehemiah,
with much preponderance of */). Thus in Ezra there are 20 instances of
*/. as against 4 of 1/7.
6. It is premature to locate the period in the history of the Hebrew
alphabet to which these interchanges may attest. For the Vorlage of
the LXX of Samuel, Driver, Samuel, lxiv thinks of ‘an early form of the
square character,’ while for the base text of the LXX of 1-2 Chronicles,
INTERCHANGES OF CONSONANTS 311
Allen, Chronicles: 2:162-65 speaks about the semicursive script of the
middle of the second century BCE. I.D. Miller tried to prove that the
interchanges between the MT and LXX in Hosea attest to the fifth and
sixth centuries, but in view of the many uncertainties it remains
difficult to make any firm statements.4 In view of the lack of
distinction between waw and yod in most of the Qumran scrolls, it seems
that the books of the LXX which show a preponderance of °/}
interchanges would reflect a relatively late stage of the textual
transmission. Indeed the books for which this is the case are the late
biblical books: Ezra-Nehemiah, 1-2 Chronicles, and, interestingly
enough, also Psalms. On the other hand, all other books display earlier
stages in the development of the Hebrew script, as the interchange 7/7
is possible in both the square script and the earlier Hebrew script.
41D. Miller, The Text of Hosea: A Demonstration That Most of the Differences in Meaning
between MT, LXX, and PSH Arose in the Late Sixth and Fifth Centuries BC, unpubl. Ph.D. diss.,
Melbourne College of Divinity, 1984.
V. THE SEPTUAGINT AND THE LITERARY CRITICISM
OF THE HEBREW BIBLE
CHAPTER TWENTY-TWO
EXEGETICAL NOTES ON THE HEBREW VORLAGE OF THE
SEPTUAGINT OF JEREMIAH 27 (34)
Every book of the LXX contains data which is important for the textual
criticism of the Hebrew Bible. A few books also contain information
which is of significance for the literary criticism of these books and, in
a way, of the whole Bible. The book which contains the most extensive
information of this kind is probably the LXX of Jeremiah.!
The LXX of Jeremiah probably reflects an early edition of the
Hebrew book, to be called ed. I, differing recensionally from the later
edition of MT, to be called ed. II.* This hypothesis is based on the fact
that the LXX is shorter than MT by one seventh and that it reflects a
different text arrangement. Both issues have been the subject of much
scholarly debate, and, as in similar cases, scholars have questioned
whether the short text of the LXX stems from a deliberate shortening
by the translator(s)? or whether it is simply derived from a shorter
Hebrew text. Scholars who have accepted the former possibility*
ascribed to the translator a free approach, assuming that he shortened
his Vorlage drastically. Such an approach derived not only from a
certain understanding of the techniques used by the LXX translators but
also from the fact that these scholars did not know of Hebrew scrolls,
such as the Qumran scrolls, which differ significantly from MT. On the
other hand, scholars who accepted the latter opinion? assumed that
1 See Tov, “Jeremiah”* and TCHB, chapter 7.
Thus “Tov, “L’incidence.” In the meantime many studies have been devoted to this
topic, expressing a view pro or contra. See Dogniez, Bibliography and my summarizing article
“The Characterization of the Additional Layer of the MT in Jeremiah,” in: Erlsr 26
(forthcoming) .
3 The problem as to whether Jeremiah was rendered by one translator, two translators,
or a translator and a reviser (thus Tov, Jeremiah and Baruch) probably does not affect the
issues discussed here.
4 Especially M.G.L. Spohn, leremtas Vates, etc. (Lipsiae 1824) 1-24; K.H. Graf, Der Prophet
Jeremia (Leipzig 1862) xl-lvii.
See especially F.C. Movers, De utriusque recensionis vaticiniorum leremiae ... indole et origine
commentatio critica (Hamburg 1837); A. Scholz, Der Masorethische Text und die LXX-
316 CHAPTER TWENTY-TWO
the Hebrew Vorlage of the LXX was much shorter than MT, lacking
many words and phrases, and also whole sentences and even passages
which are found only in MT. These scholars thus expressed confidence in
the translator’s conservative approach to the Hebrew text, although
this aspect has not been stressed to any extent. Typical proponents of a
short Hebrew Vorlage are Janzen, Jeremtah, Tov, “L’incidence” and
“Jeremiah,”* and Bogaert, “De Baruch a Jérémie.” The arguments
which support the assumption that a short Hebrew text lies behind the
LXX may be summarized as follows:
(1) Short versus long texts are found elsewhere in the LXX,
especially Ezekiel, 1 Samuel 17-18, and Joshua 6, 12, and 20 (see Tov,
“Ezekiel’”*, “Samuel”*, “Joshua”’*). The clue to an understanding of
these units lies in a correct understanding of their translational
character. If a certain unit was rendered in a free fashion, translational
omissions and additions may be expected. On the other hand, if a unit
was rendered faithfully, such omissions and additions are not to be
expected. Consequently, if a faithfully rendered translation unit is
nevertheless shorter than MT, its Vorlage was probably also shorter.
The latter situation seems to apply to Jeremiah. With the exception of
passages in which the translator encountered linguistic difficulties (for
some examples, see TCU, 162-171), Jeremiah was rendered rather
faithfully,© and the prose sections of the translation may be regarded
as literal. We should thus not expect that this translator shortened his
Vorlage substantially. On the other hand, since the book of Job was
rendered in an extremely free fashion, its short Greek text must be
approached differently.”
(2) The nature of most of the elements lacking in the LXX (the
‘minuses’§) is such that they can easily be explained as additions in ed.
II (see a tentative classification of these elements in Tov, “Jeremiah”*).
(3) The additional elements (pluses) found in ed. I often do not suit
their context. This point may be recognized from an analysis of both
content and syntax—see Tov, “Jeremiah,”* section b 1.
Ubersetzung des Buches Jerermias (Regensburg 1875); A.W. Streane, The Double Text of Jeremiah
(Cambridge 1896).
For a short description of the translation technique of the LXX of Jeremiah, see A.
Scholz Der Masorethische Text (see n. 5); F. Giesebrecht, Das Buch Jeremia (HAT, 1894) xix-
XXXiv.
7 See Gerleman, Job; D.H. Gard, The Exegetical Mathod of the Greek Translator of the Book
of Job (BL Monograph Series 8; Philadelphia 1952).
This neutral term denotes both elements actually omitted and elements which were
absent from the translator’s Vorlage.
EXEGETICAL NOTES ON JEREMIAH 27 (34) 317
(4) The name of the king of Babylon is spelled in the MT of chapters
27-29 in its later spelling Nebuchadnezzar, while in the remainder of
the book it occurs in its original form Nebuchadrezzar.? Since the name
Nebuchadnezzar is lacking in all its occurrences in chapters 27-29 in the
LXX, these may be recognized as a second layer in MT.
(5) The fragments of 4QJer>4 are very similar to the underlying text
of the LXX, both in the length and in the differing arrangement of the
text (for details see DJD XV).
(6) In several instances in which the text of Jeremiah runs parallel
with that of Kings (mainly Jeremiah 52//2 Kings 24-25), the short text
of the LXX of Jeremiah is also found in 2 Kings (both in MT and in the
LXX); see Tov, “L’incidence,” 282. |
The minuses characterize the LXX of this book as a whole and this
phenomenon is taken into consideration in the evaluation of individual
instances. At the same time, the translators did omit several small
Hebrew elements such as particles, intermissions, conjunctions, and
pronouns in accordance with their feeling for style. For example, see the
discussion below of 7% (v. 2) and of 7218 any) (v. 6). Further, the
possibility of erroneous omissions by the translator or subsequent
generations is not disregarded (for an example see the discussion of vv.
13-14).
The present study is based upon the assumption that the short LXX
text of Jeremiah reflects a short Hebrew text. This hypothesis is not
proved here, but it is illustrated in chapter 27 (chapter 34 of the LXX).
In this chapter MT contains a relatively large number of pluses over
against the LXX.
This study presents a reconstruction of the Hebrew Vorlage of
chapter 27, annotated with notes relating to the character and origin of
the additions of ed. IJ.19
1. The reconstruction
The reconstruction of the Hebrew Vorlage of Jeremiah 27 is as
problematic as any other reconstruction (for the problems, see TCU,
chapter HI), but it enables a reasonable presentation of the quanti-
? Also in other details chapters 27-29 stand out from the remainder of the book,
especially with regard to their orthography (see W. Rudolph, Jeremia [3d ed.; HAT, 1968],
ad loc.).
10 For discussions of chapter 27, also of its Greek text, see E.W. Nicholson, Preaching to the
Exiles (Oxford 1970) 94-96; W. Thiel, Die deuteronomistische Redaktion von Jeremia 26—45
(WMANT 52; Neukirchen-Vluyn 1981).
318 CHAPTER TWENTY-TWO
tative differences between the two editions of Jeremiah. The
reconstruction records quantitative differences as well as qualitative
differences (different words, etc.), if only because sometimes the two
types cannot be separated.!!
2. Chapter 27 according to the LXX (ed. I) and MT (ed. II)
The reconstructed Vorlage of the LXX!4 of Jeremiah 27 (34) is presented
on the first lines, and the expanded edition of MT on the second lines (in
italics). The text of the LXX is based on Ziegler, leremias. The text of
MT is not reproduced in full, the printed words being limited to those
instances in which they differ from the LXX. These are mainly
additions of ed. II.
The notes accompanying the reconstruction refer to the character and
origin of the additions of ed. II and they raise questions with regard to
the correctness of the reconstruction, in particular on the basis of an
analysis of the translator’s techniques.!° The notes are written in the
form of a textual-exegetical commentary on Jeremiah, contributing also
to the literary criticism of that book.!4
LXX 1
MT IN NT IITA TA ATT 730 WWI? 72 OPP NDINY PVN 1
onnn mon) now (?) qwy TDR AND: 2
onnn mon nov 77 7wy 72x87 WN 7D 2 WN? TDN TT?
PR TY 722 72 287 aR Fn Ox1 OvIR J? 2X ONN2WI 3 JANIE 2
2X7 JIDY 12 720 ON) INW TD ON) ONIN 720 ON ONTIW) 3 JIN PY
WPT BR gow ang a> xa GON Wl PTX Tn WKAR T23n
WPT PX OIwWIV ONT DINID TIPPS 72D ON) IK 72D
Weg 27:12 LXX: baa qbn px may OD ANIY NN WIA
MT: INN 791 999 72n 2VD ODN AX we
12 The orthography of the reconstruction follows MT as much as possible.
The text-critical value of many small details in grammatical categories cannot be
evaluated: disharmony/harmony in the use of pronouns, nouns, verbal forms, as well as
number; see TCU, 154-162.
According to the accepted view, the book of Jeremiah is composed of three layers,
sometimes described as sources: A (authentic sayings of the prophet), B (a biographical
account) and C (a deuteronomistic layer). It is relevant to note that several elements of the
C stratum were found to be lacking in ed. I (see Tov, “L’incidence”). Below such elements
of the C stratum which are absent in ed. I. are occasionally referred to. For this purpose we
use the list of characteristic expression of the C stratum which was compiled by J. Bright,
JBL 70 (1951) 30-35. A reference such as “Bright, 14” refers to item 14 of his list.
EXEGETICAL NOTES ON JEREMIAH 27 (34) 319
ONW? APN ‘7 WON 7D WR? OFPITN 2X ON N14 WT TP
PNW? TIN DINIX “77 WON AD FON? OTITIN ON ONN 172814 777? 720
YOST DR ON WY MDX 5 OD TR DX WRN 7D
WN WOT DN) OTR ON PONT DX CINOY VDIN 5 OD" TR ON ITONN 71D
6°12 Ww wR? Ann MwA yD YA nD
INV) 6 °°YI WwW? WR? TN TOT VIN PVT IDI PINT ID FY
Pao 77 WNIT) 3 MISINT =o AN nn) «(?)
233 72D T8N172D121 TD ONT DIZINT 2D NN VND PDIN
7 a? Twa mn nx on Way?
DN? O77 2D NN 172Y77 972)? 72-7. TWAT NN Od) 7 7TIY
0°99) 0727 OP 12 7IY) NYT ONIN NY NI TY 2 (2 AIR)
WR WD2nAM CNN 18
DN7 92.2 72D IXNITDIII DN NN 172? NWN 7D20077 77 77777 8 07777)
O° 2Y PDR AyD Ann. 933 F2n YVAN nx TN RX?
NUTT UT PY F2DN ITI) VID) IW 232 720 2V2 NIK DN INN? WN
5x1 OD N72] OX wown Ox ONND9 1D One vn Wy ‘4 ON}
5N7 OD°N722 DN wawn OX ONN)9 17° ONN ON TY ‘7 ON]
O-70X OF wR OD79WD PRI OD799 9X1 OD-N?N 2X) AD"?
DDON DDN OF WR ODDWD 2X) OD7Y 3X) ODNDIN ON) ODP
OONS PNA jy? OD? OXI) OF AWPw 7D 10292 72m MX AYN RX?
DONN P77 JY? DD? OND) OT Pw 7D 10 222 720 DN ITIVN XD WN
292 NYS NX ND WR 7) 11 o2nnIN 2yn
202 19N7K DN N72? WRK 77711 ONTIN) ODNX NTT) DINDIN Dyn
5X1 12 72 aw) 7a WOIN 2V Yon Mavi 223 72n
2X7 12 772 JW) T7129) 77 ON] INDIN PY PNA ITI) 232 72D
320 CHAPTER TWENTY-TWO
ODS NN Wea Wx? TaN oat Yop wnat AT 499 PPT
OD°INIS DN NPT WON? TON OPI TT DD NIT ATT 79D i777 718
13 yay
2972 JOY) ANN wnven 7? 13: PM) WV) INN 17DY7 732 72D 7V2
14
14.932 720 DX 12Y7 XN? WRK NT ON 77 927 WRI 7277) 272
929 72n nKx
222 720 DN VTIYN NX? WON? ODN OOK NIG IIT ON WYN ONY
apw> -ow2 ora) om ‘7 ON onn>w xX? -°D 15 002 OXI) OF APw 7D
9209 MW OND) O07 OND ONT NX? 7D 15 OD? OND OF WW "D
nD? 7pw apw(?) OD? ONDA OND ONX ONIN) OINN IT Wyn?
05? O°NI37 O°NI17) ONX ONTINI OINN 7171777 yy?
Sx ovnwn Ox “TAMER WD Tex> cna ondn 2x) aI on YD ON. 16
Sx Down ON 7 WN 7D WN> onI27 AIT OVA DOD OX) OTD ON) 16
5299 cow 47 m9 759 Fn tnx O09 ofxDIT OD NDI 7927
ANU TIOIID DIO: AN 8 AT ON 9 COND DD RS 3
17 0D? NI ANA Ww 7D
73.2 72D DN VTDY OFAN DWN ON 17 OI? DN ADT WwW D177
vy? ON) OF OND ONT18 oonndw xd
vw? ON) OF OND] ON 18 ADT ONT PVA AD? PD
"2 ND WWID? OnK ‘4707
Yona? oon 075977 2 71nd05 MND “7D 82 YD" OnK ‘7727
“Wax 7D 73 19
O°T0Y.7 ON MND 7 JON 7D 2D 19 7792 OFWITI1 TTT? 72D 17)
N AWK 20 arya ane (2) 28
NON 20 nN PYD ONT ODA An obyr miner by Oo by)
EXEGETICAL NOTES ON JEREMIAH 27 (34) 321
Tino TN ino 79770 onp»?
Td 2D fa ID? RIN AD. 929 97 WRI 912)?
o>owinen
TONS PION II 2) BOP ATT a OS RID OWI TD
22
22 OWI) ATV 720 MDT PD DPN O°7D7 VY INTW? TIN
‘7 ON! ere |
On2winy om byi7) “7 OND OMX 7PD OY TY VT AD) NIV 7722
TT QYPDT ON
1.] The original heading of this chapter may have mentioned
Zedekiah instead of Jehoiakim (Zedekiah of S represents an ancient
correction), so that the original title has either been corrupted or lost.
See further H. Schmidt, “Das Datum der Ereignisse von Jer 27 und 28,”
ZAW 39 (1931) 138-144, who claimed that the original text of 27:1
mentioned the seventh year of Zedekiah. However, probably at one
time this chapter, like several other ones, had no heading (see the
complete or partial lack of a title in ed. I in chapters 2, 7, 16, 25, 47, and
50), and the present heading was added in ed. II. The episode described
in this chapter took place during Zedekiah’s reign (see vv. 3 and 12)
and v. 1 erroneously repeats the heading of the preceding chapter
(26:1). N°77 71wI in 28:1 (lacking in ed. I) probably presupposes 27:1 in
ed. II. For a more detailed discussion of the historical background of
chapters 27-28, see A. Malamat, VTSup 28 (1975) 135, and the litera-
ture quoted there.
2. °>x ] This word was added in ed. II to the phrase ‘thus said the
Lord’ also in 13:1, 17:19, and 25:15 (as well as in ed. J in 19:1). Similar
additions are found in the Qeré text of Ruth 3:5, 17. Possibly the
pronoun was added to stress the dramatic character of the action
described here (cf. the use of the same formula in Isa 8:1, 5, 11 and Jer
18:5; 24:4).
7> ] A literal representation of this word would not have suited the
character of the Greek language, for which reason it may have been
omitted here. Similarly 7? and 05? have not been represented in the
LXX Of Gew 1251 2:47 22:2 40-19) 22:5 05) aw; 2743 2 na; Josh:20:2 tin
S322 CHAPTER TWENTY-TWO
02?; Jer 31:21 7? -a°xn. However, in general the LXX translators
represented 7? and 05° literally when used as a ‘dativus ethicus’ (e.g.,
Deut 1:13; Josh 18:4; Judg 20:7; 2 Sam 2:21). It is therefore hard to tell
whether the present omission of 7? resulted from a shorter Hebrew
Vorlage or from an omission by the translator.
3. 0°282n ] To this word the Greek translator added the pronoun
auTuv, which probably does not represent a variant reading (see n. 13).
The same applies to the omission of the pronominal suffix of onnn in
the LXX. The Lucianic tradition also omits the suffix of onn>w).
onx 7p? ] ele dmdvmmovwv abtav of the LXX (atta of manuscripts S 26
710 is probably secondary) reflects onxip? (cf. 2 Sam 5:23 n2yn xd
rendered by ox dvafyjoet ele cuvdvTyow abtay = onx zp? 72yNn xX? and cf.
a similar addition in the parallel verse 1 Chr 14:14). Alternatively, it
is unlikely that the LXX reflects 0771)? even though this word was
rendered five times by el¢c dtdvtmo.v in Chronicles: in these instances
19) denotes ‘towards’ in war contexts (see 1 Chr 14:8) or was thus
understood by the translators, but such a meaning could not be ascribed
to 1? in the present context had it appeared here.
The short reading of MT is probably more original than that of the
LXX. For additional examples of a long text of ed. I as opposed to a
short text of ed. II, see 1:17,18; 3:18; 6:16; 7:7; 14:7, 13, 15; 31:14 and
further G.C. Workman, The Text of Jeremiah (Edinburgh 1889) 70 ff.,
and Janzen, Jeremiah, 63-65.
4, x28 ] This word occurs 19 times in the MT of Jeremiah in the
phrase MxX2¥ ‘7 Ans 7D. In four of its occurrences in this phrase is mxax
reflected in the LXX, but in the remaining 15 cases it is not represented.
mixax also lacks in the LXX when used in similar expressions; for details
see Janzen, Jeremiah, 75. A case of special interest is the long phrase
ON W? 728 niNIEX ‘7, which occurs here and in another 31 verses in MT,
but never in the LXX (see Bright, 35). mx28 was thus often added in ed.
II. On the other hand, A. Rofé, “The Name YHWH SEBA'OT and the
Shorter Recension of Jeremiah,” in: R. Liwak and S. Wagner (eds.),
Prophetie und geschichtliche Wirklichkeit im alten Israel (Stuttgart
1991) 307-315, claims that this word was systematically removed from
the MT of Jeremiah, as the phrase nixlx ‘7 was invented only at the end
of the period of the Judges, and does not occur even once in Genesis-
Judges.
5. yaR7—oTRT mx ] The LXX’s omission may have resulted from
homoioteleuton if the scribe’s eye jumped from the first occurrence of
yuxn to its second occurrence. However, it is more likely that this
EXEGETICAL NOTES ON JEREMIAH 27 (34) 3Z3
section was added in ed. II: the addition is found between two segments
mwin wit DAA ona. / paxn nx cm-wy which must be taken as one
phrase in view of such verses as 32:17 puXn nN) GnwA Nk Mwy AN In
AyD JID YI N32, 10:12 and 51:15. Further, the pronominal suffix of
mnni) in v. 5b refers to the first occurrence of 7x7, a fact which makes it
unlikely that the section lacking in the LXX was omitted by mistake.
The phrase 79727 N81 O7N7 NX occurs frequently in the C stratum of
Jeremiah, see Bright, 32. For the idea expressed in the expanded text,
see Isa 45:12-13. See also M. Weinfeld apud S. Paul, Proceedings of the
Fifth World Congress of Jewish Studies ... 3-11 August 1969, I (Jerusalem
1972) 111 on the relationship between Jer 27:5 and the quoted verses of
Isaiah; A. van der Kooij, “Jeremiah 27:5-15: How do MT and LXX
Relate to Each Other?” JNSL 20 (1994) 59-78.
6. 7ny1 ] This word denotes that the speaker or author reached an
important point in a speech or discourse, but such a word could be added
at a later stage in the development of the text. This word is also absent
in ed. I in 40:4 ora Fonnnd Ain vAny) and 42:15 ‘4 727 ww 7D? Any.
Elsewhere in Jeremiah nny) was rendered faithfully by kat viv (2:18;
713) 14107 16415 2613-29-27; 32°36, 37:20; 42:227-4477). sn vr in 42:19
is represented by kat viv yudvtec yuuioecOe, i. e., WIN YT Tn) (cf. also v.
22). There was thus some textual fluidity between the two editions
with regard to this word.
"218 |] It is hard to know whether this word was found in the
translator’s Vorlage. It is represented in the LXX as part of €Swka, but
the translator could also have represented it separately, i. e., éyo
€S5wka. A similar question arises in 1:18 77nn3 13n “3x1 - [Sov TEBELKG Ge.
Tx nxaNiT ] The Vorlage of the slightly deviating translation Thy
yiiv was most likely identical with MT because similar translations are
found elsewhere in Jeremiah (see 23:3 MxqNT 2Dn - dd TaONE THe yAc;
32:37; 40:11). The translator either took mixx (countries) as meaning
‘world’ or avoided the plural form of yj (thus P. Katz, ThZ 5 [1949] 7).
*3ay/129? ] The phrase ‘Nebuchadnezzar ... my servant’ recurs in
25:9 and 43:10 where it is again absent in the LXX. Thus, Nebuchad-
nezzar is known as God’s servant in MT (ed. II) of Jeremiah, but not in
the LXX (ed. I). Some scholars believe that Jeremiah himself called
Nebuchadnezzar God’s servant and that the idea was omitted by the
Greek translator because of theological motivations (for references, see
W.E. Lemke and Z. Zevit, to be mentioned below). Of particular
interest are the discussions by T.W. Overholt, “King Nebuchadnezzar
in the Jeremiah Tradition,” CBQ 30 (1968) 39-48, and Z. Zevit, “The
Use of J1y as a Diplomatic Term in Jeremiah,” JBL 88 (1969) 74-77, who
324 CHAPTER TWENTY-TWO
describe, each in his own way, the background of 72y in the above-
mentioned three verses within the Jeremiah tradition. On the other
hand, according to W.E. Lemke, “Nebuchadnezzar, My Servant,” CBQ
28 (1966) 45-50, the mentioning of tay in MT (ed. II) derived from a
scribal error in 27:6 (see below). This verse, in turn, influenced the text
of 25:9 and 43:10. However, in our view the mentioned opinions are
imprecise because they treat the three verses on one level. However,
the problems involved in these verses are different and therefore a
middle course between the two main views may be suggested: in the two
verses in which ‘Neb. ... my servant’ is missing in ed. I, it apparently
was absent in the translator’s Vorlage, too. This is one of the many
‘omissions’ of (parts of) names in ed. I; see in particular the frequent
‘omission’ of Nebuchadnezzar’s name, as in v. 20, below. However, this
situation differs completely from the circumstances of 27:6, where
either the reading of ed. II has developed from that of ed. I, or vice
versa. Thus 72 was not added or omitted in 27:6, but it formed part of
either the original text of this verse or of a corrupted version. An
analysis of the readings can determine the way in which the corruption
went. The combined readings of ed. I and II may be recorded as 1/°71(7),
by which notation their close relationship is stressed. The added/
missing lamed in 17129? resulted by way of haplography/dittography
from the preceding 722, and the interchange of yod and waw occurs in
all stages of the Hebrew script (incidentally, a similar interchange is
found in 40:9 where MT 712y” is reflected in the LXX and in the parallel
verse 2 Kgs 25:24 as "319; cf. also Isa 66:14 r32y - Tote ceBopévoig abTtév
= 171). The graphical similarity of the two readings is better
explained by the assumption of a textual error than by a theological
change.
If indeed one reading development from the other one in 27:6, which
of the two may contextually be considered as the original? The
preferred assumption is that 112y? of ed. I is original because the
reading of ed. IT which calls Nebuchadnezzar God’s ‘servant’ is
paralleled only in two places in ed. II and these should probably be
considered as secondary. However, the reading of ed. I, 171y7, is
contextually not very plausible. First of all, the repetition of 1717 is
syntactically awkward, in particular in the short text of ed. J 771 ... "nn
Way? AWW! MeN MX Ov 722 922 72n INI7D121. Secondly, to the best of our
knowledge, elsewhere in the Bible ‘countries’ (mx7x) do not worship
God as implied by the reading of ed. I. Therefore the reading of ed. II
("J1y) probably reflects the original text which was corrupted to 172?
of ed. I, partly under influence of the ensuing 1717. At a later stage, the
EXEGETICAL NOTES ON JEREMIAH 27 (34) 325
reading of ed. II in 27:6 probably influenced the textual expansions in
25:14 and 43:10. 72 in ed. II is characteristic of the vocabulary of the C
stratum in which also David is called God’s ‘servant’ (Bright, 33). Cf.,
however, Z. Zevit, “The Use,” who explains the word as ‘vassal.’
Note further that the wording of 27:6 forms the basis of ed. II in
28:14: 177997522 459 7¥N27D722 PR Tav> 7287 OA 2D TIX DY cnn) 22-79
12 nn) 77wA Nn O47 (the italicized words are lacking in ed. I).
1? "nni ] This is probably a stylistic expansion based on the beginning
of the verse (cf. also 28:14 quoted above and a similar addition of }nx in
20:5).
7. 072173 —171291 ] The translator could conceivably have omitted this
verse prophecying submission to the grandson of Nebuchadnezzar
because, to our knowledge, Nebuchadnezzar did not have a grandson
who ruled. However, since we cannot ascribe such developed historical
motivations to the translator elsewhere in Jeremiah, it is doubtful that
they should be ascribed to him here. For the same reason it is also
unlikely that the translator would have omitted this verse as
disagreeing with the idea of an exile lasting seventy years, foretold in
Jer 29:10. Since the translator probably did not omit this verse, it must
have been lacking in his Vorlage, as suggested, too, by our general view
of the shorter text of the LXX. The idea that Babylon, the instrument of
God’s punishment, would ultimately be punished is found in additions
in both this verse and in 25:14 - lacking in ed. I— Awynd) OOVDD OF? >ND?w1
o-517) oPD1 OI OF 797 O} 02 Way *D oT -. There are also additional
parallels in both wording and content between chapter 27 and the MT of
25:8-14. The secondary character of these additions is particularly
evident in 27:7 where the added section does not conform with its
immediate context. Here nations are rebuked and warned that they are
to be punished by Babylon and in this context a punishment of Babylon
itself is not expected which will impart a completely different
dimension to the text. Further, the idea of the ultimate punishment of
Babylon is also expressed in the prophecy on Babylon (50:29; 51:24, 56)
which is generally believed to be secondary, either wholly or in part.
Finally, it should be asked whether the editor of ed. II did at all
refer to the grandson of Nebuchadnezzar or whether instead he was
using a general expression denoting subsequent generations. The
possibilities are discussed by Janzen, Jeremiah, 101-103 and Weinfeld,
Deuteronomy, 144, n. 5. If the editor of ed. II added the phrase ’... and
his son and the son of his son ...’ retrospectively, the section may have
been written after 539, the last year of Nebunaid, although in fact he
was not of Nebuchadnezzar’s offspring. According to another inter-
326 CHAPTER TWENTY-TWO
pretation, the section may have been written before 560, in which year
Evil Merodach was murdered. According to J. Bright, JBL 70 (1951) 24,
Jer 27:7 would hardly have been formulated in its present form after 560
(for possible further indications of the date of the addition in MT see
below on v. 18 ff). However, more likely is the view of M. Weiss, x7?07
inm7> (3d ed.; Jerusalem 1987) 106-110, who asserts that the phrase is
meant as a type of superlative, referring to ‘many generations’ after
Nebuchadnezzar (cf. especially Jer 2:9).
8. 7°(1) | The addition later in the verse may have led the editor of
ed. Il to expand the vague conditional waw in 717 to the fuller 1777) (cf.
also 25:12 in ed. II). V (autem) should be considered a reformulation of
MT, here and elsewhere (e.g., Jer 31:28; Isa 3:24; 7:22, 23).
wx NX+—112y" xX? ] This stylistic addition is meant to stress beyond v.
6 that Nebuchadnezzar is the instrument of punishment used by God. A
similar addition is found in 25:9 where the editor of ed. I stated in a
general way that a people coming from the North will cause a
destruction, while ed. II explicitly mentions Nebuchadnezzar. See
further the addition of ed. II in 21:7 172) 222 720 7¥N772133 772... JON
OwD1 -wPan 772) OF°2°x (the italicized words were added in ed. II).
All the prophecies in chapters 4-8 which refer to the people coming
from the North (4:5-8, 12-13, 6:1-8, 22-26, 8:14-17) mention neither
Nebuchadnezzar nor Babylon. This implies that at the beginning of his
career Jeremiah spoke only in a general way of a people coming from
the North. Babylon’s task in the punishment of Israel was mentioned
for the first time in 605 when the events had made it clear to the
prophet that the nation which God had been speaking of was in fact
Babylon: see 36:1-2, 29; 25:1-14 (MT) as well as later prophecies (cf. Y.
Kaufmann, m?x 2w77 aKxT nI71n, part 7 [Tel Aviv 1962] 404-405 and
esp. n. 7).
717121 ] The short text was expanded in accordance with the full
formula (cf. Bright, 27). Similar expansions of this formula are found in
the MT of 21:9, 44:13, and 42:17, 22 compared with v. 16.
TINIE NX yM - EuBdAwou TOV TedxnAov auTuv | The change from the
singular form of the verb to a plural one in the LXX (cf. n. 13) follows
that of its subject (‘the nation or kingdom’). |
NIT 717 99 |] This phrase was added for the sake of clarity. Similar
expansions are found often in ed. II, see, e.g., 28:12 LXX NIN, MT ANI8
NIT mT, 29:32 LXX oda1na, MT am ayn qn; 52:8 LXX nx (= 2 Kgs 25:5
LXX and MT), MT 177778 nx. See further Janzen, Jeremiah, 73-74.
EXEGETICAL NOTES ON JEREMIAH 27 (34) S27.
Onx ~an ay - €we éxAltwotv | Active verbal forms have also elsewhere
been changed to passive ones, or vice-versa (see n. 13). *»n appears very
rarely in the Bible as a transitive verb, as it does here.
9. aa-nn>n | taiv évuTrviaCopévww butv of the LXX (your dreamers), also
reflected in the other versions, may reflect 03°»?n. On the other hand,
this rendering may also reflect o>°nn?n of MT if this word was taken as
a nomen agentis (thus M. Segal, Leshonenu 10 (5699) 154-156; cf. various
other words in Jeremiah, e.g., 7713, 71N2, wp", and frequently in rabbinic
Hebrew). In that case, all translators identified o3°nn?n as the plural
(qatélét) of the nomen agentis. This assumption presupposes a devel-
oped linguistic understanding on the part of all translators (Q23°nn?n in
29:8 has not been understood in this way [Ta évitrvia vusiv], but that
word occurs in a different construction). As a consequence, if is more
likely that the translations of this verse reflect a variant 03°09n.
nx? | This word was often added in ed. II (see, e.g., 1:4; 39:16; 40:15;
45:1). The addition in the present verse may have been derived from v.
14 where 79x? 05°>X occur in a similar context: WX? O3°7X Dx ONIN
AYN NP.
10. onqax) o5nx *nni11 | The addition is based upon v. 15, a verse
which is similar in content to v. 10. The expression is characteristic of
the C stratum (see Bright, 31).
11. ‘7 ox] | This and similar phrases (‘7 7178 ON}, etc.) occur 109 fimes,
both in editions I and II. In an additional 65 instances the phrase occurs
only in ed. II.
m2) ] The translator apparently vocalized this word as 77241
(172).
12. 222 72n %y2 ] This phrase has been added from vv. 8, 11. After the
first verb in 12b was expanded with these words, the object of the
second verb had to be reduced to ‘him.’ For the phrase 03°781¥ nx& wan
ay of ed. I, cf. Neh 3:5 oF7°398 NAVA ONS wes.
522 (14) — an? (13) rm ] This long ‘omission’ of the LXX is instructive
for an understanding of the methodological problems raised by the
shorter text of the LXX. At least part of MT must be original as the next
verse makes no sense without this text. In the LXX ‘for they are prophe-
sying falsely to you’ (at the end of v. 14) refer to the king of Babylon,
but in MT they correctly refer to the false prophets mentioned in v. 14.
Therefore v. 14, now lacking in the LXX, must have been original. Hence
the translator’s omission of the section between 1729) in v. 12 and ayn
in v. 14 was probably due to homoioteleuton. At the same time, the
content of v. 13 is secondary, and we may therefore have to treat this
328 CHAPTER TWENTY-TWO
verse as other verses of similar nature are treated, as having been
absent in the LXX’s Vorlage. The added section does not mention any
new data, as it is based on v. 8 whose elements it contains in a different
order. Its secondary nature also comes to light from the phrase in the
beginning of v. 13 (‘why should you and your people die’) since it
repeats the preceding phrase ‘and you shall live’ in different words.
Ed. II thus contains both in wv. 12-13 and in v. 17 the word 1m coupled
with a rhetorical question (cf. also Ezek 18:32, 33:11). For a different
analysis of vv. 12 ff. in the LXX, see H. Seebass, ZAW 82 (1970) 449 ff.
15. *m-15 |] The pronominal suffix is not expressed in the LXX nor in V.
For the assumption of a variant 1°77, see n. 13.
o-x°ait } The LXX freely added the pronoun vtpobv (cf. n. 13).
037—o57 |] The LXX reflects a doublet 027 7pw/7pw(?) 057 on which
see Ziegler, Beitrage, 96 (our punctuation of the Greek differs from that
of Ziegler: vbutv [ev d&lkw] wevdr [13] vutv cal ...). The doublet is by
definition secondary, and the two parts of the doublet differ in the
sequence of their constituents.
For the added 7pw cf. the LXX of 14:15 and 29:23; see further above on
27:3. The assumption of a homoioteleuton 03? - 03° is less likely
because the reconstructed Vorlage of the LXX is contextually difficult.
16. ] Cf. 28:5 oy 9D cry) /oinsa cry. In the LXX translation of this
verse the order of the two phrases is inverted, as in 28:5.
T17 anv ] This addition is apparently based on the date mentioned
in 18:3 (‘two years’). Ed. I refers to the question of whether the temple
vessels will return at all, whereas ed. II raises the question of when
they will return. The phrase ‘two years’ has been added on the basis of
28:3 also in ed. II, and not in ed. I, of 28:11. According to others, these
words were omitted by the translator in accord with Bar 1:8.
17. 227n —ivnwn Ox ] The short text of ed. Tin v. 17 099 Ox) ANA Apw °D
onn>w x> is paralleled by oonn>w x) 7219 o> o-xa1 On TPpw 7D in vv. 14-15
a fact which supports our reconstruction (see further 29:9 o°x21 O71 Ww °D
o-nn?w x? ‘nwa O54). The addition in ed. II is based on 27:12 (cf. also
25:18 and 26:9). The second part of the addition is phrased as a
rhetorical question similar to v. 13.
18. “n2/72 |] In the whole section, God is mentioned in both the first
and third person. Therefore possibly "1 has been changed in one of the
traditions to ‘72 or vice versa. Alternatively, one reading may have
developed from the other on the textual level: a scribe may have
written “"2 as an abbreviated tetragrammaton which was later
misunderstood as *2, or vice versa. On the practice of abbreviating the
EXEGETICAL NOTES ON JEREMIAH 27 (34) 329
tetragrammaton, see TCHB, 256-57. Similar problems arise in 6:11 nan
min reflected in the LXX as “non; 8:14 7107 reflected in the LXX as 1;
and 40:3 4.779 onxon reflected in the LXX as 1? onxon.
71222—nin2x ] From here to the end of the chapter MT is greatly
expanded. Except for two significant additions, the expanded text
stresses details which were already found in the short text. It is
remarkable how well the editor of ed. IT managed to insert the new
elements (sometimes whole sentences) between the existing parts of ed.
I without introducing significant changes.
The author of the additions showed a great interest in the fate of
the temple vessels, adding details which are based, among other
things, on data mentioned in both Jeremiah and 2 Kings.
In the course of his reworking, the editor of ed. IT used the expression
ona o7797 (18:21) instead of the similar phrase 07797 1m found in ed.
I. These vessels (0"7D) were specified as ‘the vessels left in the house of
the Lord’ (both ed. J and II) and ‘the vessels in the house of the king’
(ed. II only). In the second detail, ed. II contains a little piece of
information not contained in ed. I which is probably reliable. In 52:13
Nebuzaradan is said to have burnt both ‘the house of the Lord and the
house of the king,’ and, as it is known that Nebuchadnezzar took
vessels from the ‘house of the Lord’ before is was burnt, he probably
acted similarly with regard to the vessels found in the ‘house of the
king.’
19. nxt Wy. on ... m1Dn7—mxax | According to ed. I, the prophet
threatened that the vessels still left in the temple would eventually be
exiled to Babylon. These vessels are specified in ed. II as: (1) the
temple vessels described here as ‘the pillars, the sea and the stands’—
this information derives from 52:17 (+ 2 Kgs 25:13) where these items
are mentioned in a different sequence; (2) ‘the rest of the vessels which
are left in this city’ (v. 19)—these are the vessels left in the royal
palace as appears from ed. II in vv. 18 and 21, even though the phrase
used in v. 19 is more encompassing. Notably, in his rephrasing of the
text, the editor of ed. II used 1m differently from its use in ed. I. In ed. I
o-927 1m denote all the vessels except for those ‘which ... the king of
Babylon did not take away’ (20), but in ed. II they refer to all the
vessels except for ‘the pillars, the sea and the stands’ (v. 19).
The reconstruction of 99 in 07237 Am %y is problematical. While in
the reconstructed ed. I these words continue the opening formula AD 75
‘7 1nx, the translator started a new sentence with them: kal tTudv étAol-
TwVY okevav (as for the remaining vessels ...). His Vorlage actually may
not have contained vy even though it is included in the full formula °5
330 CHAPTER TWENTY-TWO
2Y (MIX2xX) ‘7 WX 13 occurring in v. 21 and elsewhere in the MT and LXX
(cf. 22:6, 23:2, 15). H. Seebass, ZAW 82 (1970) 415, n. 16, reconstructed
the LXX as 1m». In principle étlAottoc may reflect both n° and o-nn1n
because the Greek word renders both words in the LXX. However, the
assumption of a condensed translation is unlikely because it disregards
the problem of the other two words which are not represented in the
LXX (9x15 7°y2).
20. on?? ] The pronominal suffix of the verb is not represented in the
LXX. This ‘omission’ may or may not represent a variant reading, cf. n.
13.
73N179121 ] Nebuchadnezzar’s name was often added in ed. I to the
phrase ‘king of Babylon,’ see 28:14; 29:3, 21; 32:28; 46:13; 49:30; 50:17.
am 72n apm 72] One of the characteristic features of ed. II is its
frequent expansion of proper nouns by adding the name of the father
and/or the title ‘king (of Judah).’ Jechoniah’s name was expanded in
this way here and also in 28:4. For similar examples of expanded names
see Janzen, Jeremiah, 139-54.
1722 ] This is an explanatory addition as in 29:4 o2wy7n cn7717 AWRK
7722,
n2wi771—nx1 ] These words were added in ed. II on the basis of a
Hebrew tradition of 29:2 underlying the LXX in which, among other
things, "1n7 is mentioned (cf. J. Ziegler, Beztrdige, 92). Cf. further 39:6
and II Kgs 24:14.
21. o2w10"—7D 7D ] This is a typical stylistic addition which neither
contains new information nor stresses any particular matter. The editor
of ed. II added so many elements in the preceding two verses that he
felt obliged to repeat parts of vv. 18-19 by way of ‘Wiederaufnahme.’
22. 7I7—orn7yT ... ONX—7nw1 ] The addition in this verse stresses
that the vessels which were still left in the temple would be exiled to
Babylon and subsequently would be returned to Jerusalem. The latter
idea is not consistent with the spirit of the surrounding verses that deal
with false prophets and not with the fate of the temple vessels. Even if
the latter would have been the case, it nevertheless seems anticlimac-
tic to have mentioned immediately after the threat to the vessels that
ultimately they would be returned to Jerusalem. The added section must
be considered secondary because of its contents and, hence regarded as a
post-exilic retrospective gloss (cf. Ezra 1:7, 11, 6:5 and Dan 5:2-3 with
regard to their wording and content). Its date may be applied to the
whole of ed. II (see also on v. 7 above).
EXEGETICAL NOTES ON JEREMIAH 27 (34) 331
°7p5 ] A similar use of this verb is found in ed. II (not ed. I) in 32:5 Jy
nx -qWpd. As a rule, this verb refers to human beings and not to inanimate
things as here. For a discussion of the uses of 1/75, see J. Scharbert, “Das
Verbum PQD in der Theologie des Alten Testaments,” BZ NF 4 (1960)
209-26.
o-na-wii | This word occurs eight times in the Bible, of which seven
are to be found in Jeremiah.
CHAPTER TWENTY-THREE
THE COMPOSITION OF 1 SAMUEL 16-18 IN LIGHT OF THE
SEPTUAGINT
In 1 Samuel 16—18—the story of the encounter of David and Goliath and
its aftermath—the LXX differs greatly from MT,! lacking 39 of the 88
verses of these chapters.* Previous discussions of these verses by
Wellhausen, Peters (see n. 2), Stoebe, and McCarter? focused on the
larger minuses of the LXX, thus neglecting three other aspects of the
LXX without which that translation cannot be evaluated well:
1. In addition to the large minuses, the LXX lacks 24 shorter elements
in these chapters, ranging from one to five words (see appendix A).
2. The LXX reflects several variants (see appendix B).
3. The LXX contains 17 pluses, ranging from single words to complete
sentences (see appendix C).
1. Approaches to the origin of the short version
The opinions expressed about the origin of the LXX’s short version of 1
Samuel 16-18 can be divided into two groups. Some scholars ascribed
the divergences between the two texts to the Greek translator, who
omitted, they claimed, 44 percent of the text because of exegetical
! The oldest attestation of the short text of the LXX is in Hippolytus' Sermo (2d century
CE) in its omission of 1 Sam 17:55-58. See the edition of G. Garitte, Traités d’Hippolyte sur
David et Goliath etc. (CSCO 263-264, Scriptores Iberici, t. 15-16; Louvain 1965). The earliest
witness of the long form of MT is 1Q7, published by D. Barthélemy in DJD I. This fragment
contains 1 Sam 18:17-18 lacking in the LXX.
2 The following verses are lacking in the OG: 17:12-31, 41, 48b, 50, 55-58; 18:1-6a, 10-11,
12b, 17-19, 21b, 29b-30. These amount to 44 percent of the verses of MT of these chapters.
We should note that whereas the OG contained in manuscripts B etc., omits these verses,
manuscripts A, etc., include a translation, which has been recognized as Hexaplaric; see R.
Peters, Beitriige zur Text- und Literarkritik sowte zur Erkldrung der Biicher Samuel (Freiburg im
Breisgau 1899) 37-38; Wellhausen, Samuel, 104; Driver, Samuel, 140; B. Johnson, Die
hexaplarische Rezension des 1 Samuelbuches der Septuaginta (STL 22; Lund 1963) 118-123. See
further n. 2 in the original article.
3 HJ. Stoebe, “Die Goliathperikope 1 Sam. XVII.1-XVIII.5 und die Textform der
Septuaginta,” VT 4 (1954) 397-413; McCarter, Samuel.
334 CHAPTER TWENTY-THREE
motives, namely, to create a smoother story by omitting conflicting
details.4 These scholars focused on the large minuses, usually
disregarding the pluses of the translation, and if they did discuss the
pluses (as did Barthélemy, for example), they also regarded them as
exegetical. According to the other, diametrically opposed view, the
LXX was based on a short Hebrew text which did not contain the so-
called minuses of the LXX.° This shorter Hebrew text was usually
considered to reflect an earlier stage of the literary development of the
story, one which preceded MT.®
It seems that no solid arguments for any one view have so far been
presented. Those scholars who suggested that the translator abridged
MT were probably influenced by the lack of supporting evidence for the
alternative explanation. Writing before the discovery of the Qumran
scrolls, they were unaware of Hebrew texts which departed as much
from MT as would the reconstructed short Vorlage of the LXX. They
therefore assumed that the shorter text was produced by the Greek
translator. The alternative view, likewise, was based mainly on
intuition and a negative judgment concerning the abridgment theory;
some of its exponents stressed that the translator was not likely to omit
such large sections and that he therefore probably found a short
Hebrew text in front of him.
4 Thus Kuenen, Historisch-kritische Einleitung in die Biicher des Alten Testaments, I, 2
(Leipzig 1890) 61; K. Budde, Die Biicher Richter und Samuel (Giessen 1890) 212; J. Schmid,
Septuagintageschichtliche Studien zum 1. Samuelbuch (Breslau 1941) 118; D. Barthélemy, “La
qualité du Texte Massorétique de Samuel,” in E. Tov (ed.), The Hebrew and Greek Texts of
Samuel, 1980 Proceedings IOSCS, Vienna (Jerusalem 1980) 1-44, esp. 17-20. The midrashic
tendencies were stressed by Barthélemy and Gooding in D. Barthélemy, D.W. Gooding, J.
Lust, and E. Tov, The Story of David and Goliath, Textual and Literary Criticism, Papers of a
Joint Venture (OBO 73; Fribourg /Géttingen 1986), as well as by A. van der Kooij, “The Story
of David and Goliath—The Early History of Its Text,” ETL LXVII (1992) 118-131.
> Thus O. Thenius, Die Biicher Samuels (Leipzig 1842) 67 (with bibliography); Peters,
Beitrage, 30-62; Wellhausen, Samuel, 105 (however, in his later Die Composition des
Hexateuchs und der historischen Biicher des Alten Testaments [3rd ed.; Berlin 1899] 247 his
attitude to the short text is unclear); H.P. Smith, A Critical and Exegetical Commentary on the
Books of Samuel (ICC; Edinburgh 1899) 150; K. Steuernagel, Lehrbuch der Einleitung in das Alte
Testament (Tubingen 1912) 317; N.C. Habel, Literary Criticism of the Old Testament
(Philadelphia 1971) 10-11; F.H. Woods, “The Light Shown by the Septuagint Version,” in:
S.R. Driver and others (eds.), Studia Biblica 1 (Oxford 1885) 21-38; Stoebe, “Goliath-
perikope”; Johnson, Rezension; McCarter, I Samuel. For a reconstruction of the original short
Hebrew text of the story, more or less identical with the Hebrew text underlying the LXX,
see Peters, Bertrage.
It is probably unrealistic to assume that some of the large minuses were due to the
translator, while others were already in his Hebrew parent text.
1 SAMUEL 16-18 335
2. Methodology
The point of departure for a new analysis must be the recognition that
the translation of 1 Samuel 17-18 has to be studied as a whole and that
any solution suggested should take into account, not only minuses, which
provide no clues for a solution, but also pluses, variant readings, and
translation technique. The inclusion of all relevant textual features
will result in a more complete and satisfactory analysis.
The idea behind such an analysis is the conviction that a translation is
internally consistent with regard to its general approach to the source
text, to which the translator is either faithful or not. If the translator
omitted 44 percent of the text, he must have approached that text
freely, and this free approach should also be visible in other details.
If, on the other hand, there are indications that the translation is
literal, that the translator approached the source text with care and
introduced but little exegesis of his own, it is not likely that he would
have omitted large sections because of exegetical (e.g., harmonistic)
motives; in that case, the short text of the LXX would more likely
reflect a short Hebrew text. These suppositions reflect a logical
inference from the act of translating, but they can also be supported by
some evidence from the translations themselves. Known Greek
translators who took care to represent the Hebrew source text exactly
showed their careful approach in all details, that is, they introduced
as little exegesis as possible in the translation equivalents and
produced a literal translation which was quantitatively equal to the
Hebrew source text (that is, without additions and omissions). This
applies to the so-called revisers of the LXX (except for Lucian) and,
within the canon of the ‘LXX,’ to the sections ascribed to kaige-Th,
Qohelet, Psalms, and, to a lesser degree, several other units as well. By
the same token, free translators show their approach to the text in
many details in the translation, for example, in their word choices and
in free additions and omissions as well as in exegetical alterations of
various types.
As a consequence, when studying the background of 1 Samuel 17-18
one should also pay attention to the translation techniques of the larger
unit in which these chapters are found,’ and in fact of the other books
7 The larger unit comprises at least 1 Samuel 1-31, but probably also 2 Sam 1:1-11:1; thus
modern scholarship in the wake of Barthélemy, Devanciers, 36 ff. According to Shenkel,
Chronology, 117-120, this unit ends at 2 Sam 10:1; according to B.H. Kelly, The Septuagint
Translators of I Samuel and II Samuel 1:1-11:1, unpubl. diss. Princeton Theological Seminary
1948, it ends at 2 Samuel 5.
336 CHAPTER TWENTY-THREE
of the LXX as well. But the main focus remains the character of these
two chapters.
3. The texts
A full reconstruction of the Hebrew Vorlage of the LXX would
unnecessarily complicate the present study (for an attempt, see Peters,
Beitrdge). For our purposes it suffices to present a translation of the MT
of 1 Sam 16:17-18:30, indicating where the LXX differs from it.2 The
narrative shared by the LXX and MT is printed in Roman type. Points
at which the LXX shows minor deviations from MT, where the LXX
probably reflects different readings (see Appendix B), are indicated by
underlining. Elements which are absent in the LXX (small minuses) are
indicated by parentheses (see Appendix A). Small pluses of the LXX
are not indicated here (see Appendix C), nor are exegetical renderings
reflecting the translator’s exegesis. Portions of the narrative found only
in MT are printed in italics.
16:17 So Saul said to his courtiers, “Find me someone who can play well
and bring him to me.” !8One of the attendants spoke up, “I have observed a
son of Jesse the Bethlehemite who is skilled in music; he is a stalwart felow
and a warrior, sensible in speech, and handsome in appearance, and the
Lord is with him.” 19Whereupon Saul sent messengers to Jesse to say, “Send
me your son David, who is with the flock.” 2°Jesse took an ass laden with
bread, a skin of wine, and a kid, and sent them to Saul by his son David.
21So David came to Saul and entered his service; Saul took a strong liking
to him and made him one of his arms-bearers. 22Saul sent word to Jesse,
“Let David remain in my service, for | am pleased with him.” “Whenever
the [evil] spirit of God came upon Saul, David would take the lyre and play
it; Saul would find relief and feel better, and the evil spirit would leave
him.
17:1The Philistines assembled their forces for battle; they massed at
Socoh of Judah, and encamped at Ephes-dammim, between Socoh and
Azekah. “Saul and the men of Israel massed and encamped in the valley of
Elah. They drew up their line of battle against the Philistines, 3with the
Philistines stationed on one hill and Israel stationed on the opposite hill;
the ravine was between them. 4A champion of the Philistine forces stepped
forward; his name was Goliath of Gath, and he was six cubits and a span
tall. "He had a (bronze) helmet on his head, and wore a breastplate of scale
8 The translation follows NJPST, with minor adjustments; words in sqaure brackets are
explanatory editions of the NJPST translators. The text of the LXX follows codex B.
XP ty
1 SAMUEL 16-18 337
armor, a bronze breastplate weighing five thousand shekels. "He had
bronze greaves on his legs and a bronze javelin slung from his shoulders.
7The shaft of his spear was like a weaver’s bar, and the iron head of his
spear weighed six hundred shekels; and the shield-bearer marched in front
of him.
8He stopped and called out to the ranks of Israel and he said to them,
“Why should you come out to engage in battle? I am the Philistine
champion, and you are Saul’s servants. Choose one of your men and let him
come down against me. If he bests me in combat and kills me, we will
become your slaves; but if I best (him) and kill him, you shall be our slaves
and serve us.” !9And the Philistine ended, “I herewith defy the ranks of
Israel. Get me a man and let’s fight it out!” !41When Saul and all Israel
heard these words of the Philistine, they were dismayed and terror
stricken.
12David was the son of a certain Ephrathite of Bethlehem in Judah whose
name was Jesse. He had eight sons, and in the days of Saul the man was
already old, advanced in years. 1>The three oldest sons of Jesse had left and
gone with Saul to the war. The names of his three sons who had gone to the
war were Eliab the firstborn, the next Abinadab, and the third Shammah;
14Qnd David was the youngest. The three oldest had followed Saul, Sand
David would go back and forth from attending on Saul to shepherd his
father’s flock at Bethlehem.
16The Philistine stepped forward morning and evening and took his
stand for forty days.
17}esse said to his son David, “Take an ephah of this parched corn and
these ten loaves of bread for your brothers in camp. 18T ake these ten cheeses to
the captain of their thousand. Find out how your brothers are and bring some
token from them.” 19Saul and the brothers and all the men of Israel were in the
valley of Elah, in the war against the Philistines.
20Farly next morning, David left someone in charge of the flock, took [the
provisions], and set out, as his father Jesse had instructed him. He reached the
barricade as the army was going out to the battle lines shouting the war cry.
21 Israel and the Philistines drew up their battle lines opposite each other.
22David left his baggage with the man in charge of the baggage and ran
toward the battle line and went to greet his brothers. @3While he was talking
to them, the champion, whose name was Goliath, the Philistine of Gath,
stepped forward from the Philistine ranks and spoke the same words as
before; and David heard him.
24When the men of Israel saw the man, they fled in terror. 25 And the men
of Israel were saying, “Do you see that man coming out? He comes out to defy
Israel! The man who kills him will be rewarded by the king with great riches;
338
CHAPTER TWENTY-THREE
he will also give him his daughter in marriage and grant exemption to his
father’s house in Israel.” 26David asked the man standing near him, “What
will be done for the man who kills that Philistine and removes the disgrace
from Israel? Who is that uncircumcised Philistine that he dares defy the
ranks of the living God?” *’The troops told him in the same words what
would be done for the man who killed him.
28When Eliab, his oldest brother, heard him speaking to the men, Eliab
became angry with David and said, “Why did you come down here, and with
whom did you leave those few sheep in the wilderness? I know your
impudence and your impertinence: you came down to watch the fighting!”
29But David replied, “What have I done now? I was only asking!” 28And he
turned away from him toward someone else; he asked the same question, and
the troops gave him the same answer as before. 31 The things David said were
overheard and were reported to Saul, who had him brought over.
32David said to Saul, “Let no man’s courage fail him. Your servant
will go and fight (that) Philistine!” 33But Saul said to David, “You cannot
go to that Philistine and fight him; you are only a boy, and he has been a
warrior from his youth!” *4David replied to Saul, “Your servant has been
tending his father’s sheep, and if a lion or a bear came and carried off an
animal from the flock, 9°I would go after it and fight it and rescue it from its
mouth. And if it attacked me, I would seize it by the beard and strike it
down and kill it. 2°>Your servant has killed both lion and bear; and (that)
uncircumcised Philistine shall end up like one of them, for he has defied the
ranks of the living God. 37The Lord,” (David went on,) “who saved me
from lion and bear will also save me from that Philistine.” “Then go,” Saul
said to David, “and may the Lord be with you!”
38Saul clothed David in his own garment; he placed a bronze helmet on
his head (and fastened a breastplate on him). 9?David girded his sword
over his garment. Then he fried to walk; but he was not used to it. And
David said to Saul, “I cannot walk in these, for 1am not used to them.” So
he (David) took them off. #He took his stick, picked a few smooth stones
from the wadi, put them in the pocket of his shepherd’s bag and, sling in
hand, he went toward the Philistine.
41 The Philistine, meanwhile, was coming closer to David, preceded by
his shield bearer. 42(And the Philistine looked) and he saw David; he
scorned him, for he was but a boy, ruddy and handsome. #*And the
Philistine called out to David, “Am 1 a dog that you come against me with
sticks?” The Philistine cursed David by his gods; *4and the Philistine said
to David, “Come here, and I will give your flesh to the birds of the sky and
the beasts of the field.”
1 SAMUEL 16-18 339
4°David replied to the Philistine, “You come against me with sword
and spear and javelin; but J come against you in the name of the Lord of
Hosts, the God of the ranks of Israel, whom you have defied. 46This (very)
day the Lord will deliver you into my hands. [ will kill you and cut off
your head; and I will give the carcasses of the Philistine camp to the birds
of the sky and the beasts of the earth. All the earth shall know that there is a
God in Israel. 4”7And this whole assembly shall know that the Lord can
give victory without sword or spear. For the battle is the Lord’s, and He
will deliver you into our hands.”
48When the Philistine began to come (and advance) toward David,
David quickly ran up to the battle line to face the Philistine. 4?David put his
hand into the bag; he took out a stone and slung it. It struck the Philistine in
the forehead; the stone sank into his forehead, and he fell face down on the
ground. °°Thus David bested the Philistine with sling and stone; he struck
him down and killed him. David had no sword. °4So David ran up and stood
over the Philistine, grasped his sword (and pulled it from its sheath); and
(with it) he dispatched him and cut off his head.
When the Philistines saw that their warrior was dead, they ran. -2The
men of Israel and Judah rose up with a war cry and they pursued the
Philistines all the way to Gai and up to the gates of Ekron; the Philistines
fell mortally wounded along the road to Shaarim up to Gath and Ekron.
“3Then the Israelites returned from chasing the Philistines and looted their
camp.
“4David took the head of the Philistine and brought it to Jerusalem; and
he put his weapon in his own tent.
2°When Saul saw David going out to assault the Philistine, he asked his
army commander Abner, “Whose son is that boy, Abner?” And Abner
replied, “By your life, Your Majesty, I do not know.” °6”Then find out whose
son that young fellow 1s,” the king ordered. 57So when David returned after
killing the Philistine, Abner took him and brought him to Saul, with the head
of the Philistine still in his hand. “8Saul said to him, “Whose son are you, my
boy? And David answered, “The son of your servant Jesse the Bethlehemite.”
18:lWhen he finished speaking with Saul, Jonathan’s soul became bound up
with the soul of David; Jonathan loved David as himself. 2Saul took him
[into his service] that day and would not let him return to his father’s
house.—?Jonathan and David made a pact, because he loved him as himself.
4Jonathan took off the cloak and tunic he was wearing and gave them to
David, together with his sword, bow, and belt. “David went out, and he
was successful in every mission on which Saul sent him, and Saul put him
in command of all the soldiers; this pleased all the troops and Saul’s
courtiers as well. 'When they came home [and] David returned from killing
340
CHAPTER TWENTY-THREE
the Philistine, the women of all the towns of Israel came out (singing and
dancing to greet King Saul) with timbrels, shouting, and sistrums. 7The
women sang as they danced, and they chanted: Saul has slain his
thousands; David, his tens of thousands! 8(Saul was much distressed) and
greatly vexed about the matter. For he said, “To David they have given tens
of thousands, and to me they have given thousands. (All that he lacks is the
kingship!).” ?From that day on Saul kept a jealous eye on David. !The next
day an evil spirit of God gripped Saul and he began to rave in the house,
while David was playing [the lyre], as he did daily. Saul had a spear in his
hand, 14and Saul threw the spear, thinking to pin David to the wall. But
David eluded him twice. }2Saul was afraid of David, for the Lord was with
him and had turned away from Saul. 13So Saul removed him from his
presence and appointed him chief of a thousand, to march at the head of the
troops. 14David was successful in all his undertakings, for the Lord was
with him; !°and when Saul saw that he was successful, he dreaded him.
16A)1 Israel and Judah loved David, for he marched at their head.
17Saul said to David, “Here is my older daughter Merab, I will give her
to you in marriage; in return, you be my warrior and fight the battles of the
Lord.” Saul thought: “Let not my hand strike him; let the hand of the
Philistines strike him.” \8David replied to Saul, “Who am I and what is my
life—my father’s family in Israel—that I should become Your Majesty’s son-
in-law?” 1?But at the time that Merab, daughter of Saul, should have been
given to David, she was given in marriage to Adriel the Meholathite. 2°Now
Michal, daughter of Saul, had fallen in love with David; and when this
was reported to Saul, it (the matter) was pleasing for him. 7/Saul thought:
“T will give her to him, and she can serve as a snare for him, so that the
Philistines may kill him.” So Saul said to David, “You can become my son-in-
law even now through the second one.” 2And Saul instructed his courtiers
to say to David privately, “The king is fond of you and all his courtiers
like you. So why not become the king’s son-in-law?” *2When the king’s
courtiers repeated these words to David, David replied, “Do you think that
becoming the son-in-law of a king is a small matter, when I am but a poor
man of no consequence?” “4Saul’s courtiers reported to him (saying), “This
is what David answered.” *°And Saul said, “Say this to David: ‘The king
desires no other bride price than the foreskins of a hundred Philistines, as
vengeance on the king’s enemies.’” —Saul intended to bring about David's
death at the hands of the Philistines —2°When his courtiers told this to
David, David was pleased with the idea of becoming the king’s son-in-law.
(Before the time had expired,) 2”David went out with his men and killed
two hundred Philistines, (David) brought their foreskins (and they were
counted out) for the king, that he might become the king’s son-in-law. He
1 SAMUEL 16-18 341
(Saul) then gave him his daughter Michal in Marriage. 28When Saul saw
(and knew) the Lord was with David and that Michal daughter of
loved him, 2?and he (Saul) grew still more afraid of David; and Saul was
David's enemy ever after.
30The Philistine chiefs marched out to battle; and every time they
marched out, David was more successful than all the other officers of Saul.
His reputation soared.
4. Translation technique
Five aspects of the LXX are analyzed here: (1) hnguistic versus
exegetical renderings; (2) word order; (3) quantitative representation;
(4) consistency in translation equivalents; (5) Hebraisms in the
translation. These five aspects of translation technique are suitable for
testing the relative degree of literalism or freedom with which the
translator approached the Hebrew text. The analysis shows that the
translator of 1 Samuel 17-18 remained relatively faithful to the
Hebrew text, and it is therefore unlikely that he would have omitted
44 percent of that text. In other words, the LXX was based on a short
Hebrew text containing only that part of the story presently found in
the LXX (as well as in the corresponding verses in MT); the remaining
material, now found only in MT, was not included in that short text.
a. Linguistic versus exegetical rendering 9
Technically a distinction between ‘linguistic’ and ‘exegetical’ render-
ings is a bit misleading, in that this terminology implies that linguistic
renderings are not exegetical. Actually, even a linguistic rendering
reflects exegesis, though of a strictly technical type.!9 The following
list contains examples of contextual-exegetical renderings (in some cases
the possibility of a variant reading [indicated by !] is not excluded):
17:2 ORIw wer) Kal ol &vSpec Iopand
and the men (lit. man) of and the men (pl.) of Israel
Israel
172° SAAR Paya !év TH KOLAGSL. abTol Tapa-
TA OOOVTAL
in the valley of Elah in the valley.
and they drew up battle lines They drew up battle lines
1722. <n O11) Tapatdooovtat el¢ 16\ELOV
(they drew up) battle lines they drew up a line for battle
9 For the theoretical background see Tov, TCU, 50 ff.
10 For a detailed analysis, see pp. 107-108 in the original article.
342
LAS
75
17:7
17:7
17:8
17:8
17:9
17:9
17:9
17:9
17:34
17:35
17:35
17:42
17:46
17:47
18:14
CHAPTER TWENTY-THREE
abate)
were stationed (pl.)
Teawi (772w01)
(and the weight of) the
breastplate
ein naw
and the head of his spear
En
the shield
manda ays
to draw up battle lines
-nwan
the Philistine
“nx onda (227 on)
(if he is able) to fight with
me
497
and smites me
19 DDN 78 ON)
lit., but if Iam able to him
Ana)
and you will be
No)
and there came
oP?"
and he rose up
yam rnDsmM
and I struck him and killed
him
"I)TN)
and ruddy
Sew? OTN we
there is a God fo Israel
(In BAPTA 2D) WIA}
and they shall know
> a eee ie,
lit., to all his ways
loTatar
was stationed (sg.)
Kal 6 otaOudc Tob bwpakoc
avTou
and the weight of his
breastplate
Kal f} ASyxT avuTot
and his spear-head (one
word)
ta StrAa avtov
his shield
TapaTdéaoGat ToXE LW
to arrange yourself for battle
add HvdA0G
(an [no article]) alien
Kal éav Suvi] tpdc Eye
TOE LT|OaL
if he is able with me to fight
Kal ¢av TaTdén LE
and if he smites me
lédv 6€ €ya) Suvned
but if I am able
€aeabe
you will be
Kal dtav HpYETo
and when there came
Kal el €travlotato
and if he rose up
Kal étrdtaéa kal é6avdtwoa
avuTdov
and I struck and killed him
Kal AUTOG TUPPAKNG
! and he (was) ruddy
! €otiv 8edcg Ev Iopana.
! there is a God in Israel
!kal YVWOETAL
! and he shall know
!é€v tdoatc taic dbotc abtov
!in all his ways
1 SAMUEL 16-18 343
16:20: “I !kal dtmmyyéaAn
and they (pl.) reported and if was reported
18:23 TPN Kal ovxl EvSoEog
and unimportant and not important
In analyzing chapters 17-18 we are interested in forming a judgment on
the amount of exegetical renderings the translation contains. The above
list shows that these chapters contain only a limited amount of such
exegesis (at most 22 examples in 17 of the 49 verses present in the LXX),
especially if one takes into consideration that some nine of the
deviations listed may reflect variant readings.
b. Word order
With the exception of 17:9, -mx an?a? 237% (‘shall be able to fight with
me’) vs. S5uvnér mpdc Eye Todepoat (‘is able against me to fight’), the
translator kept the exact word order of MT. The differences in word
order in 17:38 and 18:7, 22 (twice) probably derived from a different
Hebrew text.
c. Quantitative representation
Partly as a result of the tendency toward stereotyping, literal trans-
lators did their utmost to represent each individual element in MT by
one equivalent element in the translation. Free translators, on the other
hand, felt free to add clarifying elements or not to represent elements
which, in their view, were expressed by other words in the translation.
They often compressed two or more elements of the Hebrew text into
one, and expanded one element into two or more, in accordance with
their literary taste and the nature of the Greek language. The
quantitative relationship between the source text and the translation
can be expressed statistically. The more literal translators aimed at a
one-to-one representation of words in MT, whereas free translators did
not.
The LXX translation of 1 Samuel 17-18 usually follows a system of
precise quantitative adherence to the Hebrew. Some exceptions, which
partially overlap with the list of exegetical elements in the trans-
lation (above), are listed here:
17:7) an ta StAa avtot
the-shield his shield
17:9 71Dm Kal €av TaTdéy pe
and-strikes-me and if he strikes me
1737 ANN aT Kal 4 AdyxN abtod
and the head of his spear and the spear-head of his
344 CHAPTER TWENTY-THREE
but if I am able to him but if Iam able
17:9, ann €aeabe
and you will be you will be
17:34 KX Kal Stav tpxeTo
and there came and when there came
L735 “O27 Kal el €tavlatato
and he rose up and if he rose up
18:23. FPN Kal ovyl €vSoEoc
and unimportant and not important
d. Consistency in translation equivalents
Many translators rendered all occurrences of a give Hebrew word,
element (e.g., preposition), root, or construction as far as possible by the
same Greek equivalent, often disregarding the context and the effect of
this type of translation on the quality of the translation. There are two
aspects to such consistency: (a) internal consistency in the choice of
translation equivalents within a certain textual unit and (b) the
translator’s adherence to the general vocabulary of the LXX. No firm
data for the comparison of 1 Samuel 17-18 with other translation units
are available, so we must content ourselves with mere impressions. It
seems that in the matter of consistency 1 Samuel 17-18 reflects a type of
translation which holds the middle ground between literal and free
translations.
e. Internal consistency
Most translation equivalents in 1 Samuel 17-18 are internally
consistent, that is, the translator used the same equivalent for words
which occur in more than one place. For example:
NON ouvd' yw collect 17:1, 1,2
mhtata) Tape LBoAy camp 17:2, 46
mn Tape wba AAW encamp 17:1;.2
Wy TapaTdoow draw up battle 17:2,8
lines
yn1D/? Tepikedadkata helmet 17:5;.50
najoqvy» tapataé.c¢ ranks 17:8, 10, 36, 45
(also 17:4)
nan ave rdi Cw defy 17:10, 36, 45
x7 PoBé oar fear 17:11, 18:12
re KG SLOV wallet 17:40, 49
Ww- evOUve be set right 18:20, 26
Lack of consistency is visible in the following equivalents:
1 SAMUEL 16-18 345
4ay do00XA0G slave 17:9, 9, 32, 34
Tatc servant 18:22, 22, 23, 24
2a éxkoTmrdw) rescue 17:35
€Earpéw 17:37
227 Baxtnpla stick 17:40
b&BS0c 17:43
The differentiation may be intentional as Goliath calls David’s Baxtnpla
(staff) a mere paBSoc (stick).
non TAaTACOW strike 17:9, 35, 49; 18:6, 27
TUTTW 17:36
dtoktelvw 17:46
f. Adherence to the general vocabulary of the LXX
The basis of the vocabulary of the LXX was established by the
translators of the Torah. The translators who translated the later
books often adhered to this vocabulary, certainly the more literal ones
(see Tov, “Pentateuch”*). Thus S€tpo and etcaoSog (see below) are words
that would not usually be chosen as equivalents for the Hebrew words
they render. The examples mentioned in the preceding section as well
as the following ones reflect this approach:
Pa dvd jLeéecov between passim
poy KOLAdG valley 172
yaw Bupa€ breastplate 17:5, 5
my oa shield 77
Mon7n wee = avijp man of war 17233
TTOAELLOTHG
27 dtrep( TH NTOG uncircumcised 17:36
7D? Go! Sedpo Come! 17:44
ili? éxkAnola assembly 17:47
yon ddahdCw cry out 17:52
7N2 €L050G all the way to 1752
vIn oKdvdadov snare 18:21
ipalasarn éTUyayuBpe vw become related by — 18:22, 23, 26, 27
marriage
mx EVTEAAOLLAL command 18:22
Wy dxpoBvotia foreskin 18:25)27
ANN dyatdw love 18:16,.20; 22, 28
Tan atayyé Aw report 18:20, 24
Unusual word choices, not (or rarely) used elsewhere in the LXX, are
found in the following:
346 CHAPTER TWENTY-THREE
okie) OKéAN legs (usually: 168€c) 17:6
"199 on tTpoTopevouat walkin front (usually 17:7
two words)
on? Wovonaxéovat fight (usually: 17:10
TOAELLEW aS in VV. 32,
33)
Wm duddote pou together (usually: dua) 17:10
It seems that the translation equivalents used in 1 Samuel 17-18 reflect
a rather consistently Septuagintal type of translation.
g. Hebraisms in the translation
On the basis of the above data, the translation technique of 1 Samuel
17-18 may be described as relatively literal. A similar conclusion has
been reached by others with regard to 1 Samuel as a whole.!! Special
mention should be made of Sollamo, Semiprepositions, esp. 280 ff.
which yielded the conclusion that 1 Samuel belongs to the most literal
units of the whole LXX. On the basis of a similar study by Soisalon-
Soininen, Infinitive, esp. 169 ff., 1 Samuel may be characterized as
relatively literal. Two types of data support this characterization.
Numerous Hebraisms appearing in the translation illustrate the
translator’s literalism. In the following these are in italics.
Led Mw 727A Ww 772 uN
And they encamped between Socoh and between Azekah.
Kal TapeuBdddovo.y dvd péoov Zokyw6 kal dva péoov ACnka
17:4 wend)... Oa wee ANN
And a champion stepped forward ... Goliath (was) his name.
Kal €€yjev dvnp Suvatoc...Pod.ab dvoya avras
17D WII? NIT OwPwP 7rAW)
And with armor of scales he was dressed.
Kal 8ubpaka ad\vorSwTdov avTog EvdESuKiC
17:9 4") 0k 23 OR
If he is able ... then (lit. and) we will become.
11 Thus Thenius, Biicher Samuels, xxv ff.; Woods, Light, 21; Driver, Judaean Scrolls, lix-Ixii,
with many examples. Likewise Kelly (cited in n. 7), 24 (’... which aim at literalism to a
greater extent than the majority of the Septuagint books’), though the greater part of
Kelly’s study discusses the translator’s exegetical deviations. The predominantly exegetical
character of the translation is maintained in a brief study by Gehman, “Exegetical
Methods,” 292-296. However, the issue is not whether there are exegetical renderings in
the LXX of 1 Samuel—the existence of some of these is apparent—but how many are
found in that translation unit when compared with its literal renderings. In our view
exegetical renderings are much less frequent than literal renderings. Further, many (most?)
of the examples can also be explained as reflecting variant readings.
1 SAMUEL 16-18 347
Kal €av Suvnfy ... Kal €adépeba
17:33- “Onna NI??
... to go... fo fight
Tope vOVaL... TOU TOAE WEtV
17:40 19 Jwx ova °2D2 onx aw
And he put them in the shepherd’s bag which he had.
Kal €8€To avTotc év TH kablw Tu TOLLEVLKG) TH 6vTL aUTU
17:42 9 AX 75° OV (ITN) WI)
He was a boy, ruddy with beauty of appearance.
autoc fv tatSdpiov kal atitég TuppaKns peta KdAAOUG bday
17:43 nv29n27ON ND ANN
You come against me with (lit. im) sticks.
ov Epxn év Eve év PdBbw (reflecting variant 2021, with a stick)
For a similar use of év, see vv. 43b, 45, 47; 18:6.
18:8 WW W377 rPysya
And this matter (word) was evil in his eyes.
Kal Tovnpov Eddavn TO pha év dfOadpoic Zaovad trepl TOV Adyou
TOUTOU
And the word was evil in the eyes of Saul concerning this word.
For similar constructions, see 18:20, 23, 26.
18:12) 717.71D?n DINw RW
Saul was afraid from the face of David.
Kal E€moBHhEyn UaovaA did mpocutrov Aave.d
18:22 42nq 72 pon
the king is fond of (lit., in) you
6 Bactrevc bédXet Ev col (cf. also v. 25)
18:27 9 PWIND NIT IAI
And he went out, he and his men.
Kal €TopevOn avtoc Kal ol dvdpec avtou
18:27) o-nw>n2 79
literally: And he smote in the Philistines.
Kal €tdtatev €v Totg dhAomvAoic
Hebraisms in the pluses (not found in MT) underscore the translator’s
adherence to his parent text:
17:8 é& évavttac Faiv
= WNN723
to meet us
17:36 obyl topetvoopat kal TaTadEw avTov kal adders otpepov SvEeldoc
= FIN ara nyo PND 728 N27
Shall I not go and smite him and remove today disgrace?
348 CHAPTER TWENTY-THREE
17:48 el¢ ovvdvtTno. Aauid
= 77 nx
... to meet David.
Note further the use of \éywv (= 7987) in a plus in 18:22.
h. The argument from translation technique
The above-mentioned data show that the translator remained, as a
rule, loyal to his parent text, and it is therefore not likely that he
would have omitted 44 percent of the text. We therefore assume that
the translator worked from a text which was much shorter than MT.
This working hypothesis is supported by three arguments:
1. Confidence in the reliability of the LXX of Samuel has been
enhanced in recent years by the finds of Hebrew scrolls of Samuel in
Qumran. These scrolls contain many readings which had been
reconstructed previously from the LXX (either the mainstream or
LXXLuc). This situation thus gives the LXX more credibility in those
chapters of which no ancient Hebrew manuscripts have been found. At
the same time, the differences between MT and the reconstructed parent
text of the LXX are larger in 1 Samuel 17-18 than in any other section of
the book;!* nor do any of the Qumran scrolls differ as much from MT.
The only parallels showing similarly extensive divergence from MT
which come to mind are the large plus of 4QSam? before the beginning
of 1 Samuel 11 (five lines) and the beginning of the second column of the
same scroll (1 Sam 2:13 ff.), which differs considerably from MT.!8
2. The working hypothesis, that the short version of the story found
in the LXX is based on a short Hebrew original, is more acceptable if
the alternative view, that it is an abridgment by the Greek translator,
cannot be sustained. Indeed, in our view there are no cogent reasons for
assuming a large-scale shortening of the original text by the translator.
One might suppose, for example, that the translator omitted a
substantial portion of the narrative in order to shorten the lengthy
stories. But the argument from translation technique militates against
this supposition: The translator has not shown himself willing to take
such liberties with his source elsewhere. Furthermore, the presence of
pluses in the translation also gainsays such an assumption.
12 Elsewhere in 1 Samuel the LXX lacks individual phrases or clauses, but nowhere does
it lack so many as in chapters 17-18. For some examples, see 1:9; 4:17; 6:4, 11; 10:16; 12:13;
21:10; 23:23; 26:4; 30:7b; 31:6. For a discussion, see Méritan, La version grecque des livres de
Samuel (Paris 1898) 139-48.
13 See Cross, “Ammonite Oppression” (see p. 293); idem, “New Qumran Fragment.”
1 SAMUEL 16-18 349
3. The motive usually given to explain why the translator would
have abridged is that he recognized difficulties in certain passages,
which he therefore omitted. Two examples of such difficulties are the
following:
a. In 17:55-58, Saul and Abner express ignorance of David when they
see him approaching Goliath, and Saul asks to have David introduced
to him. This contradicts the scene preceding the battle, where Saul and
David have a lengthy discussion about David’s confronting Goliath
(17:31-39), and the earlier story of David’s being introduced to Saul as
a skilful harper and being made his armor bearer, where it is even said
that Saul ‘loved’ David (16:17-23). It is often claimed that the
translator omitted 17:55-58 to eliminate this contradiction of the
earlier scenes.
b. In 18:17-19, Saul offers David his eldest daughter, Merab, while
verses 20-26 tell about David’s marriage to Michal, ‘daughter of Saul’
(vv. 20, 27). The tension between these passages is apparent (despite
the harmonizing remark in v. 21b), and this may have promoted the
translator to omit the first section (vv. 17-19), which is now lacking in
the LXX. See further section 5.
That a translator omitted complete sections from his parent text to
avoid inconsistencies is a legitimate assumption, albeit a very difficult
one. It presupposes not only that the translator allowed himself
considerable liberty in his translation, but also that he was a
sophisticated reader, almost a critical scholar. It is questionable
whether there are any parallels for such a presumed action within the
realm of the Greek translations of the Bible. Scores of contradictory
passages have been left everywhere else in the translation, including
the LXX of Samuel (see section 5). Not only is the mere fact of the
omission surprising, so is the assumed reason for that omission, which
ascribes to the translator the mind of an attentive critic.
More important, while a harmonizing omission by the translator in
the above two examples is, in view of their contents, at least plausible,
such an assumption is much more difficult, if not impossible, in the case
of the other minuses in the LXX. In 18:1-4 we are informed of the
covenant of love between David and Jonathan; why should that section
be omitted? And why should verses 5—6a, which merely introduce the
next section, be omitted? True, 18:1-6a too contains a detail which could
be read as inconsistent with the earlier narrative: In 18:2 Saul installs
David in his court, even though he had already been installed there in
16:22. But should we expect the translator to be sensitive to such
details? And even if we should, why should the translator omit six and
350 CHAPTER TWENTY-THREE
a half verses because of one detail (18:2)? Would it not have been easier
and more responsible merely to change a detail (e.g., in 17:15) or to omit
a smaller part of the section in question? Did the translator omit 18:10-
11 (Saul’s attempt to spear David) because it is repeated in 19:9-10? Or
did he consider this section inconsistent with Saul’s feelings of love for
David? The latter possibility is unlikely, because the translation also
lacks 18:2a, which mentions Saul’s love.
The same types of questions may be asked regarding the translator’s
supposed omission of 17:12-31, the largest of the minuses of the LXX in 1
Samuel 17-18. This section contains several elements that contradict
the preceding or following account (see section 5), but all these
contradictions are relatively minor, and we do not know whether the
translator would have sensed them. But even if he would have, would a
translator omit a complete section of twenty verses because of
difficulties regarding some of the verses in that section?
Apart from these questions, two other considerations show the
inadequacy of harmonization as an explanation for the minuses in the
LXX of 1 Samuel 17-18. First, several of the minuses show no
inconsistency with the remaining text, and there would have been no
reason to omit them on that score (17:41, 48b, 50; 18:12b, 29b-30). And
second, not all difficulties have been removed from the version found in
the LXX: 17:33, in which David is called a mere lad, unqualified to
fight Goliath, remains, despite its apparent inconsis-tency with 16:18,
where he is called a man of valor and a man of war (see. n. 18).
In sum, we cannot think of any motive which would convincingly
explain an abridgment of the text. Only in a few cases can one point to
possible reasons for a stylistic or exegetical abridgment of individual
passages, and these are not sufficient to establish a case for extensive
abridgment. These considerations also militate against the likelihood
that the short text was the result of abridgment by a Hebrew scribe
(rather than the Greek translator), as suggested by A. Kuenen. Such a
theory would encounter the same objections as those just discussed, as
well as another: It is highly unlikely that the Hebrew text would be
revised only in chapters 17-18 and not in other chapters in 1 Samuel
which contain obvious contradictions and doublets of stories (e.g., the
different traditions concerning the origin of the monarchy in 1 Sam 8:1-
22; 10:17-27 // 9:1-10:16; the parallel stories about David and Saul in 1
Sam 19:11-17 // 19:18-24 // 20:142; 1 Sam 24 // 1 Sam 26).
1 SAMUEL 16-18 351
5. The two versions underlying 1 Samuel 17-18
What emerges from the preceding discussion is that the short version of
1 Samuel 17-18 reflected in the LXX was not an abridgment, either by
the Greek translator or by a Hebrew scribe, of the long version found in
MT. It is rather an independent and coherent version of the events. In
what follows we analyze the nature of this version and its counterpart
in the passages absent from the LXX and found only in MT. In so doing,
we turn from the realm of textual criticism to that of literary criticism.
The argument up to this point implies that the short version
underlying the LXX reflects an early stage of chapters 17-18 (continuing
chapter 16 [see n. 14]) and that the long version found in MT represents a
later, expanded stage. Since the long version contains additional
information (traditions) about the encounter of David and Goliath,
parallel to that in the short version, the additional material in the
long version constitutes a separate version of the story. We refer to the
short text underlying the LXX (and parts of MT) as version 1 and the
additions found only in MT as version 2.14 MT thus contains both
versions 1 and 2.!° In a way, this situation resembles that in Jeremiah
where a short edition of the book is contained in the LXX and 4QJer>.4
and a long one in MT (see Tov, “Jeremiah’*).
For a more detailed analysis we present a summary of the contents of
the two versions, disregarding small pluses and minuses.!®
Version 1 (LXX and MT) Version 2 (MT only)
16:17-23 David is introduced to Saul
as a skilful harper and he
is made his armor bearer.
7a Attack by the Philistines.
Goliath suggests a duel with
14 Version 1 is taken as reflecting the main story of 1 Samuel (i.e., it follows chapter 16
and continues with chapter 19), since version 2 has been superimposed on it and inserted
in it. This is a logica] inference from the relationship between versions 1 and 2, but
considering the contents of both versions, it is not impossible that version 2 also reflects the
framework of 1 Samuel (not, e.g., the depiction of David as a shepherd boy in version 2 and
in 16:11, 19).
This terminology is appropriate for the two versions of the encounter of David and
Goliath (chapter 17) and for the two versions of Saul’s offer of marriage (18:17-19, 20-27),
but not for other details in version 2, which are not parallel to version 1, but rather expand
version 1. Since the majority of the pluses of MT add parallel material, it is best to use the
term versions.
© Most commentaries merely remark on the relation between the two versions of the
story of David and Goliath, but McCarter, I Samuel presents the two versions as two
independent units (‘David and the Philistine Champion I, II’), translating and
commenting on them separately.
CHAPTER TWENTY-THREE
one of the Israelites.
17:12-31 David is sent by his
father to bring food to
his brothers at the
front. He hears Goliath
and desires to meet him
in a duel.
17:32-39 David volunteers to fight
with Goliath.
17:40-54 The duel. After Goliath’s Short account of the
miraculous fall, the duel (vv. 41, 48b, 50)
Philistines flee.
17:55-58 Saul asks who David
is. David is introduced
to Saul by Abner.
18:1-4 David and Jonathan
make a covenant.
18:5-6a David is appointed as
an officer in Saul’s
army.
18:6b-—9 Saul’s jealousy of David.
18:10-11 Saul attempts in vain
to kill David.
18:12-16 David's successes.
18:17-19 Saul offers David his
eldest daughter,
Merab.
18:20—27 Saul offers David his
daughter Michal.
19:29b-30 Saul’s love for David.
David’s successes.
The parallels between the two versions of the events are that in each
David is introduced to Saul (16:17-23 [part of an earlier section of
version 1] and 17:55-58) and that in each David is made an officer in
Saul’s army (18:5, 13). Furthermore, in each version Saul offers David
one of his daughters (both termed ‘daughter of Saul’: 18:19, 20),
without any cross reference to the offer of the other daughter (18:17-19,
20-27 |see, however, section 5, on 18:21b]). At the same time, the two
versions are not fully parallel, as they often contain different elements.
Version | is much more extensive than version 2, as is obvious from a
1 SAMUEL 16-18 303
comparison of the two accounts of the duel. Version 1 presents a
continuous!’ and internally consistent story,!8 and if version 2 were not
known, we would not have lacked any information in chapters 17 and 18
which is crucial to the understanding of version 1.!9 Whether or not
version 2 once existed in a fuller form, from which the present form was
excerpted, cannot be known.
The two versions underlying chapters 17-18 contain only partial
parallels, and because there is not sufficient evidence for contrasting
the two stories, it is unclear whether the duplication should be
connected with other duplications in Samuel. Even though several
parallel versions of events have been detected elsewhere in Samuel, it
is hard to know whether the two versions of the encounter of David and
Goliath should be connected with these other duplicate strands of
tradition.
From the point of view of literary history, we consider version 1 to be
more original than version 2, since the latter has been added to it (or,
rather, inserted in it). However this does not imply that the content of
version 1 is more authentic than that of version 2. For example, we do
not express any opinion on the type of description of David’s person
which is found in the different versions. It is hard to know whether
“David the harper and the armor bearer’ (version 1) is more original in
the history of the tradition than ‘David the shepherd’ (version 2, but
also 1 Sam 16:11, 19). The later tradition depicts David as both a
musician and a shepherd (see e.g., Psalm 151 in 11QPs? and in the LXX).
Version 1 in chapter 17 thus should not be preferred to version 2 from
the point of view of its contents. In chapter 18, at times version 1 is
preferable to version 2, and at times the mere editorial juxtaposition of
versions 1 and 2 creates contextual problems that render the isolated
reading of either version 1 and 2 desirable. This refers especially to the
17 17-32 links immediately with 17:11, not with 17:31 (29 “because of him,’ in verse 32
probably refers to Goliath, and Goliath has not been mentioned in the verses which
immediately precede verse 32 in MT, but he is mentioned in verse 11 [alternatively, yoy
means ‘upon himself’]). In the other instances too the verse in MT which immediately
precedes the minus has its natural continuation in the verse following the minus.
18 4 slight problem is created by a companiscn of 16:18 and 17:33. In the first verse,
David is described as AnMn wad Oem 723, ‘a man of valor and a man of war,’ while in the
second Saul advises David not to fight because he is a mere 1Y}, ‘lad.’ The tension between
these two verses may be misleading. It is possible that the phrase in 16:18 is an exaggeration
by one of Saul’s men; possibly he means to say that David has the right traits for a warrior.
Likewise, Saul’s statement in 17:33 could be exaggerated (cf. the use of 1¥1 in 1 Kings 3:7).
19 One difficulty is created by the covenant of friendship between David and Jonathan
mentioned in 18:14 (version 2) and subsequently referred to in 20:8. If we assume that the
redactor who joined versions 1 and 2, the latter including 18:1—-4, wrote or rewrote 20:8, the
problem is solved.
354 CHAPTER TWENTY-THREE
two versions of Saul’s offer of a daughter to David in marriage (18:17—
19 [version 2], 20-27 [version 1]) and to Saul’s attempt to kill David (vv.
10-11 [version 2]). All exegetes agree that Saul’s attempt to kill David
is not in place in this chapter (it is repeated by an identical section in
19:9-10). In fact, the sequence of events in the short version 1 is more
logical than that in the combined text of versions 1 and 2. In version 1,
Saul is at first envious of David (vv. 8-9), then suspicious (v. 12) and
frightened because of David’s successes (vv. 13-15); subsequently he
wants to have David killed by the Philistines, and when this
stratagem does not succeed, he attempts to kill him himself (19:9-10).
In the combined version of MT, the progressive intensification of Saul’s
response is undercut by Saul’s premature attempt in 18:10-11.
6. The composition of the Masoretic version of 1 Samuel 16-18
From the above discussion it is clear that the Masoretic version of 1
Samuel 16-18 was created by the juxtaposition of the two separate
accounts of the events, the complete version 1 and the partial (or
partially preserved) version 2.
Since both versions cover some of the same events, but with differing
details, the conflate Masoretic version which was produced by the join
contains several inconsistencies:
1. The most conspicuous difficulty, as explained above, is that after
David had been introduced to Saul and had become his armor bearer
(16:17—23, from version 1), he is absent from the battle front and
occupied as a shepherd with his father’s flock and is still unknown to
Saul who, when David arrives, has to ask Abner who he is (17:55-58,
from version 2). Note that Saul asks in general terms about ‘the boy’
(57595;-50):
2. In 17:22 (the first sentence of version 2), David and Jesse are
introduced to the reader, but they were already known from chapter 16
(version 1).
3. If Eliab was present at the time of David’s anointing (16:13, from
version 1), it is hard to understand why he should utter such harsh
words to David (17:28, from version 2). If the issue is judged only ona
psychological level, it is understandable that the oldest brother might
be jealous or anxious about the safety of his youngest brother.
4. David is depicted in different ways in the composite narrative. In
16:21 he is Saul’s armor bearer (from version 1), and in that capacity he
fights Goliath. In 17:12-31 and 55-58 (from version 2), he is an unknown
shepherd boy who happens to be on the spot visiting his brothers when
Goliath challenges the Israelites to a duel.
] SAMUEL 16-18 355
5. In 18:13 (from version 1) David is made an officer in Saul’s army,
though he was already made an officer in 18:5 (from version 2). This
inconsistency holds as long as the two appointments are not taken as
referring to different positions.
6. According to 17:25 ff. (from version 2), whoever defeats Goliath is
to be given the king’s daughter in marriage. 18:20 ff. (from version 1)
seem unaware of this promise, since Saul has to look for pretexts that
would convince David to marry his daughter, while David says that
he is unworthy.
7. According to 18:20-27 (from version 1), Saul offers David Michal,
‘daughter of Saul,’ but in verses 17-19 (from version 2), Saul offered
David his eldest daughter, Merab, also termed ‘daughter of Saul,’ in
accordance with his earlier promise to marry his daughter to whoever
defeats Goliath (17:25, likewise from version 2).
The fact that the redactor who combined versions 1 and 2 created a
text displaying such inconsistencies is precisely what is supposed to
have happened in other cases throughout the Bible where texts
underwent conflation, expansion, and interpolation. Why the redactor
created this conflate version, despite its inconsistencies, is a matter of
conjecture. It stands to reason that he wanted to preserve certain
traditions and details that were not included in version 1, which
formed the framework of his story. Presumably the redactor derived
most of version 2 from a written source. It is hard to determine why he
added 17:12-31 and 55-58 (the main body of version 2). Possibly he
simply liked the story; possibly he wanted to convey a certain idea it
expresses, namely, that God can bring victory to his people even
through initially unimportant figures (in this version David was
unknown before the battle). Other additions may reflect the editor’s
own ideas.29 In verse 50, for example, he stressed that David did not
need a sword in order to defeat the Philistine.
Still, the redactor did not necessarily ignore all the inconsistencies
created by his juxtaposition of the two versions. There are a few details
in the’ text which have the effect of smoothing out certain of the
inconsistencies. If we did not have the evidence of the LXX that the
narrative is indeed composite, we might take such details as evidence
for its original unity, but since that is ruled out, these details have
plausibly been taken as belonging to neither version but rather as
20 For further speculations on the different tendencies visible in the two versions, see
esp. Peters, Bettrige, 57; de Vries, “David’s Victory”; Jason, “Story of David and Goliath.”
According to the latter, version 1 reflects a ‘romantic epic’ and version 2 a ‘heroic epic.’
356 CHAPTER TWENTY-THREE
composed by the redactor for the purpose of smoothing out the
inconsistencies. Here are some examples:*!
a. ni, lit. “this one,’ in 17:12: ond man A cM)KX wee 12 71), ‘David
was the son of an Ephrathite man, this one, from Bethlehem.’ Since
David’s father** had already been introduced in chapter 16, his
introduction in 17:12 would have seemed repetitious and oblivious of
the earlier introduction. The ungrammatical use of the demonstrative
particle in this verse suggests that it was added by the redactor to
remove the impression of obliviousness (proper Hebrew usage would
have been 717 *MNT wenn, ‘this Ephrathite man’; the formulation w-x
*n)x, ‘an Ephrathite man,’ is correct only without 717). In context the
particle must mean ‘the aforementioned,’ as Jerome understood it (de
quo supra dictum est).*°
b. 17:15 on? ma yax 8 nx my Dxw Yun aw) 720 W711, ‘David would
go back and forth from attending on Saul to shepherd his father’s flock
at Bethlehem.’ Since David had already left him and become Saul’s
armor bearer (16:17—23, version 1), the fact that he was still with Jesse
when Saul and the army were at the front (17:12-20, version 2) would
have seemed inconsistent. 17:15 smooths out the inconsistency by indi-
cating that David alternated his time between home and Saul’s court.
c. 1 Sam 18:21b ‘you can become my son-in-law even now through the
second one’ (NJV), added in version 2, may be in the nature of a cross-
reference to the mentioning of the other daughter (Michal) in version 1.
The present study shows that the Masoretic version of 1 Samuel 16—
18 combined two originally separate versions of the narrative. The
versions sometimes told of the same incidents, though not always with
identical details; at other times they told of different incidents. As a
result, when the two versions were joined, the combined text displayed
a certain amount of redundancy and inconsistency. In a few places the
redactor added notes in an attempt to smooth over these difficulties; in
other places he made no such attempt.
The results of the analysis are of importance for our understanding
not only of 1 Samuel 16-18, but of other sections of Samuel too, and in a
way of the whole of biblical literature. In this case we are able to
document the existence of two layers of one story, while in other cases
the assumption of different layers is merely an abstract possibility.
21 For the technique and one additional example from Samuel and one from Genesis, see
Seeligmann, “Hebrdische Erzahlung,” esp. 312-314.
Even if a1 refers to David (thus Qimhi), it would still be considered an editorial or
scribal addition.
Alternatively, 717 is a corruption of 77 (interchange of zayin and yod).
1 SAMUEL 16-18
Appendix A
Shorter minuses in the LXX of 1 Samuel 17-18
The items missing in the LXX are enclosed in parentheses.
17:5
17:9
1733
17:36
17537
17:36
17:39
17:42
17:46
17:48
1751
17:51
18:6
18:7
18:8
18:8
18:20
18:24
18:26
18:27
18:27
18:27
18:28
18:29
(nwni) 97191
(19) 2D 7X ORI
(av) snw>p7
(ain) 2av7 snw?2an
(717 77N79)
(aw ine wl?)
Y 20m C117) G0"
(nw>D7 V2")
(a1) arn
(a5). 425
19n nen)
(Aqynn 75>w>)
(77a) n0971
(AaXwW MRAP MyPMAT Aw?
Jn )
(mypnwrn) Own
(aN 21RW? 77)
(AD1207 4X 12 TV)
yprya (290) Ww
(7982) 12 Dikw vay 77
(O77 INN RP)
(377) x27
(Daxw) 12 qn
qn? (01897071)
(9771) TINW NW
xq (21NwW) FON
and a (bronze) helmet
but if I am able (to him)
(this) Philistine
(this) uncircumcised Philistine
(and David said)
(and dressed him in a breastplate)
and (David) [he] took them off of
him
(when the Philistine looked)
(this) [to-]day
and went (and drew close)
and he took his sword
(and pulled it from its sheath)
and he cut off (with it)
(singing and dancing towards king
Saul)
the (dancing) women
(and Saul was greatly angered)
(and all that he lacks is the king-
ship)
and (the matter) was pleasing in
his eyes
Saul’s servants reported to him
(saying)
(Before the days were fulfilled)
and (David) [he] brought
and (Saul) [he] gave him
(and they were counted out) for the
king
and Saul saw (and knew)
And (Saul) [he] became more afraid
358 CHAPTER TWENTY-THREE
Appendix B
Variant readings reflected in the LXX of 1 Samuel 17-18
The LXX and MT readings are presented in parallel columns, with
tentative retroversions of the variants reflected in LXX added in a
third column. Differences that may be due to translation technique are
indicated with an asterisk.
LXX MT Retroverted variant
17:2 év Ti Kotdd&t. avTol «DAV APRA pov3 TX pOV3
TAPATAOOOVTAL
in the valley. They in the valley of
drew up battle lines Elah and drew up
battle lines
17:4. é€k Tie TapatdEewe =o nunnn monyvnn
from the battle line from the camps
17:4 — (txpoc adtod) (max) ww (173) VIN
Tecodpwv (mxewv)
(his height was) (his height was)
four (cubits) six (cubits)
17:7, kal 6 kovTég rm yy)
and the (wooden) and the shaft
pole
17:8 °EBpatot ody oynay
Hebrews servants
17:9 *kal édv ON ON)
and if i
17:32 Tot Kkuplou pou OTN “JAR
my lord man
17:34 kal 1 dpKkoc 377 nN} 277 4X)
and a bear and the bear (acc.)
17:35 Tov ddpvyyos ad’tot 1712 72
of his throat of his beard
17:36 Kal THY dpKkov 27777 0) 7087 OX Oh 21TH (ns) O)
(EtuTTEV 4 SotAdG (Jay 737) 79N7 (MX) OD
cou) Kal Tov A€ovTa
both bear (has your _ both lion and bear
servant killed) and (has your servant
lion killed)
17:37 *kal €otat ktiptoc Wit mM ivan
may be the Lord
(with you)
17:38 Kal (treptkedadalav)
and (a helmet)
17:39 Kal €xotlacev
and he was unable
17:39 dta€ kal &l¢
once and twice
17:40 *ele audd0yrV
into (his) bag
17:42 Kdddouc >8adyuiv
beauty of eyes
17:43 év padBSo
with a stick
*kal dtTroKAElOEL O€
and he will deliver
you
17:46 Ta Kid cou Kal Ta
17:46
KWAG
your carcasses and
the carcasses
17:47 *kal yuwoEeTal
and it will know
(all this assembly)
Kal dvéotn
and he went up
17:48
17:51 éw avtév
over him
17:52 Teé
Geth
17:52 dtlow avtuiv
after them
17:52 * AokaXuvoc
Ashkelon
17:53 dvépec lopanadA
1 SAMUEL 16-18
may the Lord be
(with you)
yap jn
and placed a
helmet
IN
and he tried
NOI
for he was not used
to them
DIP PD
and into (his) bag
ne 115°
handsome of
appearance
mp2
with sticks
70°
he will deliver
you
pa)
the carcasses
(ata DapA 3D) WIM
and they will
know
oly eae ary ais
and it happened
that he went up
“Nw7D7 ON
to/over the
Philistine
N7}
Gal
o-nw?5n n& (1977°))
(and they pursued)
the Philistines
py
Ekron
ONW? 12
yI}71
0772
Oy
Ypna
130")
>92D1 JUD
vr)
071
POR / Po"
ni
Ons
1)22WN
DNIw wer (cf. v. 2)
360
18:6
18:8
18:8
18:14
18:16
18:21
18:22
18:25
18:25
18:27
18:28
18:28
18:28
men of Israel
al xopevovcat
the dancers
év odBadoic Zaovdr
in the eyes of Saul
TEpL TOU Adyou
about the matter
*€v Tdoatc taic
dd50tc auTov
in all his
undertakings
TIpO TPOOUTIOU TOU
Aaov
before the people
él DAOvA
against Saul
Kal ov
and you
“arn 7
other than
auTov €pBadetv
to cast him
EKATOV
one hundred.
Kal 1TdG
and all
lopana
Israel
tyyatra avutéov
he loved him
CHAPTER TWENTY-THREE
sons of Israel
o win
the women
2°92
in his eyes
2
the matter
SA 257
to all his
undertakings
o77153
before them
12
against him
nny}
and now
se
than
TI AN 2979
to cast David
O NN
two hundred
257)
and Michal
DIRw NI
the daughter of
Saul
WNAAX
she loved him
Sow 3y7y2
"oT 252
ovn 755
Rw
On)
OX 3D
shista i)
rN?)
23)
Sew
JAMIN
17:5
17:8
1732
17:36
LJ2o7
17:40
17:42
17:43
17:46
17:47
1 SAMUEL 16-18
361
Appendix C
Pluses in the LXX of 1 Samuel 17-18
The majority of the pluses can be tentatively retroverted into Hebrew;
they are the elements after the plus sign or in between two plus signs in
the list. What stands outside these signs is present in MT.
LXX
xadkovd + kal oLdrpou
brass + and iron
Tokéuw + €& €vavttac Huaiv
battle + toward us
1) + 87 + oupTIEGETW
let not + I pray + fall
+ oUXL Topetooual Kal tatdéw
auTov Kal amedus oe pov
dverBog €€ Topand StdtTe Tie
6 dtepttuntog otrog +
+ shall I not go and smite him
and remove today disgrace
from Israel, for who is this
uncircumcised +
ToU add\AodvAov + Tot
ATEPLTHNTOU + TOUTOU
this + uncircumcised +
Philistine
TPOG + TOV avdpa + TOV
ddAASHvAOV
to + the man + the Philistine
Kal elSev + Todrad
and saw + Goliath (subject)
+ kal ALOoLce Kal elev Aavetd
ovxyt GAA’ 7H xelpw Kbvo> +
+ and stones and David said,
No, but rather ...+
Kal dtrokAeloet Ge KUPLOG +
OT|LLE POV
and the Lord will deliver you
+ today
Kal twapaéwoet + KUPLOG
Retroverted variant
372) + nwn
wnNap? + mann
2D” +X] +N
+ nm) VnDTM 72x x17
SS ON DUN AD Aon
in Pn
Atta 21M enw Pon
“nwoon + wont + 5x
moy+exrm
+ ORO ROTI AR
+
on + “7 N07
T+ 7M)
362
17:49
17:49
18:6
18:22
18:22
18:24
18:27
CHAPTER TWENTY-THREE
and will give + the Lord
(subject)
ALBov + Eva.
stone + one
Kal 5:é5u 6 Al60g + 8a THe
Teplkedadalac +el¢ TO
WLETWITOV auTot
and the stone penetrated +
through the helmet + into his
forehead
+ ele ovvdvtTnotv Aaverd +
+ towards David +
+ M€ywv - + saying
Aadnjoate + bet
speak + you (pl., subject
pronoun)
KaTa Ta pyuata TattTa + 4 +
éAGANOEV
according to these things
+ which + he spoke
THY Medyxoa bBvyatépa avtot +
auTa +
his daughter Michal + to him
nN + TON
+ Mel WY [Axe YI0M
WX?
+ TT ART? +
+ TWN? +
OnNN + 27
A+ SUN + APRA OAD
$47 na Dy ns
CHAPTER TWENTY-FOUR
THE LITERARY HISTORY OF THE BOOK OF JEREMIAH IN LIGHT
OF ITS TEXTUAL HISTORY
1. The two editions of Jeremiah
The LXX of Jeremiah often differs from MT of that book in major
details. It is shorter than MT by one-seventh and it differs from MT in
its arrangement of the material.! Since the LXX’s translation technique
in Jeremiah is relatively literal where the two texts overlap,? it is
unlikely that the translator would have abridged his Hebrew Vorlage.
This implies that the brevity of the LXX reflects a short Hebrew text.
The existence of a short Hebrew text of Jeremiah has been confirmed
by 4QJer®-4 containing parts of chapters 9-10, 43, and 50 (see DJD XV).
These manuscripts, dated to the first half of the second century BCE,
resemble the LXX of Jeremiah in the two major features in which the
reconstructed Vorlage of that translation differs from MT, namely, the
arrangement of the text and its shortness. 4QJer®4 share seven minuses
with the LXX, two of which are long (10:6-8, 10), and five short
(mainly names).° In addition, two minuses of the LXX are not shared
with 4QJer>-4,4 while ten short minuses cannot be compared with
4QJer>-4 because of their fragmentary nature. The reconstructed text of
4QJer® also agrees with the LXX (against MT) in the sequence of the
verses in chapter 10, where the verses appear in the order 1—5a, 9, 5b,
Pil:
At the same time, 4QJer?-4 are not identical to the reconstructed
Vorlage of the LXX. In addition to the three minuses of LXX which are
! This situation is most clearly visible in the different location of the oracles against the
foreign nations. In MT they constitute chapters 46-51, but in the LXX they follow 25:13
and are arranged in a different order.
For a description, see Min, Minuses and Pluses, and, more briefly, A. Scholz, Der
Masorethische Text und die LXX Ubersetzung des Buches Jeremias (Regensburg 1875) and F.
oe Das Buch Jeremta (HAT LI 2.1; Gottingen 1894) xix—xxxiv. See also p. 348 above.
3 43:4, 5, ‘Son of Kareah’; 43:5, ‘from all the countries to which they had been scattered’;
43:6, ‘chief of the guards’ (a°nav 17); 43:6, ‘son of ‘Shaphan.’
4 43:7, ‘land’; 43:9, ‘in mortar in the brick structure which’ (AWX 722n2 v2).
364 CHAPTER TWENTY-FOUR
not shared with 4QJer®-4, the scrolls agree with MT against the LXX in
five details, and they also contain some unique readings found in
neither the LXX nor MT. While 4QJer®-4 are thus not identical to the
Vorlage of the LXX, the existence of such a short and differently
ordered Hebrew version of Jeremiah, coupled with the fact that the
translator of Jeremiah was relatively literal and not likely to have
made such changes himself, confirms the assumption that the LXX of
Jeremiah was based on a short Hebrew Vorlage, similar to 4QJer>-4.
The question of whether the short or the long version of Jeremiah is
the earlier of the two has been discussed by Janzen, Jeremiah; Min,
Minuses and Pluses; and Tov, “Exegetical Notes.”* Each of these studies
suggested that the short version is earlier than the long one. It has been
suggested further, especially by the present writer and by P.-M.
Bogaert,> that the common text of the LXX and 4QJer>-4, that is, the
short version, does not reflect a different text of Jeremiah but an earlier
edition of that book (edition I; by the same token, MT is called edition
IT). The two editions differed from each other not only in length but
also in the arrangement of the material. The first, short edition was
expanded to the form now found in ed. II during one of the stages of the
literary growth of the book. Edition IJ contains many pluses over
against ed. I, not only in words, phrases, and sentences, but also in
complete sections, the largest of which are 33:14-26 and 39:4-13. The
date of the textual witnesses of the first edition does not bear on the
date of the edition itself, because presumably ed. I was written long
before the time of the LXX, and it was not destroyed even when ed. II
was created on the basis of ed. I. It was still known in the second century
BCE in Egypt, when it served as the Vorlage for the LXX translation,
and was present (along with manuscripts close to ed. If) at Qumran in
the first half of the second century BCE.
The description of ed. II that follows is based on the premise that
the Hebrew Vorlage of the LXX and 4QJer>-4 represent an early edition
of Jeremiah which was expanded by the editor of MT into ed. II.
° Tov, “L’incidence” and “Exegetical Notes”; Bogaert, “De Baruch a Jérémie” and
“Mécanismes.”
6 The terms edition /editor and text /scribe describe different stages in the development of
the book as well as the persons involved. Editionsbelong to the stages of the growth of the
book, up to and including its final formulation, and they involve major changes, additions,
and transpositions; the writers who produced them are termed editors. The textual
transmission, performed by scribes for each edition, starts after that edition was completed.
Scribes involved in this process did insert changes into the text, but to a much smaller
degree than editors did.
LITERARY HISTORY OF JEREMIAH 365
2. The nature of the added layer of edition II] (MT)
When inserting his own words and thoughts in a book that was
transmitted under the name of the prophet Jeremiah, editor If took
considerable liberty. Indeed, pseudepigraphal authorship and revision
were common practice in antiquity.’ Editor II did not distort
significantly the message of the prophet as handed down to him. True,
he added a great deal and inserted significant changes, but these
changes were not radical. Furthermore, editor I] did not rewrite a scroll
that contained only authentic Jeremianic utterances, but he found the
deuteronomistic edition of Jeremiah’s sayings and biography so that
much of what he added was based on an already edited book (see
below, a iil). .
The main aspects of the additions and changes of ed. II are reviewed
next. Edition II is sometimes shorter than ed. 1,8 but these relatively
few instances may be disregarded in the overall evaluation of ed. II. It
is assumed that both 4QJer®-4 and the Hebrew Vorlage of the LXX
developed from an earlier form of ed. I and that editor II rewrote a text
which was very similar to ed. I, but not identical with it.
Editor II rewrote, reedited, and revised a text almost identical with
ed. I, even though ed. I] does not reflect a consistent rewriting of the
previous edition. Revisional activity in literary compositions bears a
very personal and subjective character, and this should be borne in mind
when the additions of ed. I] are analyzed.
The anonymous editor II was not a scribe, but he produced one of the
stages of the literary work now called MT. He had access to genuine
Jeremianic material not included in ed. I, he rearranged sections, and he
also added new material (for all these, see section a). Editor If revised
an edition that was known to him in a written form as he often inserted
elements neatly between the words found in ed. I.
The inconsistency of his rewriting cannot be taken as an argument
against our working hypothesis since very few revisions are
consistent—in the biblical realm only ‘inconsistent’ revisers are known,
such as the deuteronomistic reviser of Joshua through 2 Kings, the
7 Cf. M. Smith, “Pseudepigraphy in the Israelite Literary Tradition,” in: Pseudepigrapha |
(Geneva 1972) 191-215, esp. 200 ff.
8 For the data, see Janzen, Jeremiah, 63-65 and Min, Minuses and Pluses. It is not likely that
editor II omitted these details, as they resemble typologically similar details that were
added in ed. J.
366 CHAPTER TWENTY-FOUR
‘Elohist’ in the Psalms, the Lucianic reviser of the LXX, and, on a
different level, SP.?
a. Editorial aspects
Editor II took the liberty of adding and changing many minor details
and a few major ones:
1. Addition of headings to prophecies
Editor II added several headings to prophecies which in ed. I had no
heading at all; he also expanded existing short headings:!9
2:1—2 The word of the Lord came to me, saying, Go proclaim to
Jerusalem.
7:1-2 (The word which came to Jeremiah from the Lord: Stand at the
gate of the house of the Lord, and there proclaim this word:)
Hear the word of the Lord, all you of Judah (who enter these
gates to worship the Lord).
The added information in this last heading derives from v. 10 and the parallel
passage in chapter 26.
16:1 The word of the Lord came to me.
27:1 At the beginning of the reign of King Jehoiakim son of Josiah son
of Judah, this word came to Jeremiah from the Lord.
This heading is added wrongly, for the chapter itself speaks of Zedekiah; cf. vv. 3,
12, and 28:1ff. The added heading erroneously repeats that of the previous
chapter, 26:1.
47:1 (The word of the Lord that came to the prophet Jeremiah) con-
cerning the Philistines, (before Pharaoh conquered Gaza).!!
? The inconsistency of the deuteronomistic reviser of Joshua through 2 Kings and
Jeremiah reveals itself in the discrepancies between the proto-deuteronomistic
composition and the deuteronomistic layer, and further in the amount of intervention
differing from one section to the other. Similarly, the ‘Elohist’ in the Psalms did not replace
all occurrences of 717°. The Lucianic reviser was not consistent in his vocabulary, linguistic
changes, and the degree of intervention. In SP, harmonizing editing is applied only to some
stories.
10 When the additions are quoted along with their larger contexts, parentheses are used
for the added matter. When the additions supplement pronouns which were implicit in
verbs, the pronouns are placed in brackets. For additions that are presented without any
context, no parentheses are used. The Hebrew text (reconstructed from the LXX where
necessary ) of passages quoted here can be found in Tov, “Some Aspects.” The English
translations follow NJV as much as possible.
The addition of headings in the prophecies against the nations. is inconsistent
Indeed, some ‘historical’ headings are found in editions I and II in the prophecies against
Egypt (46:1, 13), Kedar (49:28-33), and Elam (49:34-39), while other prophecies have no
heading at all in editions I and II: Moab (48), Ammon (49:1-6), Edom (49:7-27), Babel (50-
51):
LITERARY HISTORY OF JEREMIAH 367
ii. Repetition of sections
In ed. I, several sections of two or more verses occur twice, for example,
6:22-24 = 50:41-43; 10:12-16 = 51:15-19; 23:19-20 = 30:23-24; 49:18-21 =
50:44-46. This repetition may have originated with the prophet
himself, who applied certain prophecies to more than one situation or
it may have derived from editor I. Against this background editor II
felt at liberty to continue this practice. The following sections are
duplicated in ed. II: 6:13-15 (duplicated in 8:10b-12); 15:13-14 (dupli-
cated in 17:3-4); 46:27—28 (duplicated in 30:10-11); 49:22 (duplicated in
48:40b, 41b).
iii. Addition of new verses and sections
Editor II added a substantial number of new verses, both in prose and in
poetry, which derive from the three major strata recognized in the
book. Some of these are presumably authentic utterances of the prophet
(stratum A), others belong to the historical-biographical stratum (B,
by Baruch?), and still others, written in deuteronomistic diction
(stratum C), were probably composed by editor I] himself.
We first turn to stratum A in which editor II added some original
Jeremianic verses and passages. We do not know why this Jeremianic
material had not previously entered ed. I and why or how it was
preserved. One should not doubt the originality of these verses (see
below on 33:14—26) just because they were lacking in ed. I. Authentic
material probably continued to circulate among the prophet’s followers
even after ed. I was completed. For a possible parallel, see the
authentic traditions relating to 1-2 Kings contained in Chronicles.
The most remarkable addition of this kind is the prophecy in 33:14—
26 on the 7P78 Nox (‘true branch’) and the durability of the covenant.
Although this section has often been denied to Jeremiah because it is
absent from the LXX and may have been added secondarily on the basis
of 23:5-6 and 31:35-37,)4 there is no sound reason for this scepticism. On
the contrary, in addition to 33:14-16, 25-26, which resemble the
aforementioned passages, there are several Jeremianic expressions in
this section reminiscent of other passages in the book,!% and the
argument that these elements reflect a glossator’s imitation is
12 See W. Rudolph, Jeremiah (HAT; 2d ed.; Tiibingen 1958) 199-201; J. Skinner, Prophecy
and Religion (Cambridge 1963) 310. For a discussion of the whole issue, see esp. Janzen,
Jeremiah, 122-23.
13 -The good thing’ (v. 14), cf. 29:10, ‘the house of Israel ... and the house of Judah’ (v.
14), cf. 3:18; 13:11; 31:27; 31:31; ‘in those days and at that time’ (v. 15), cf. 50:4, 20.
368 CHAPTER TWENTY-FOUR
artificial. The burden of proof is on those who deny the section to the
prophet in whose name it has been transmitted.
Other sections added in poetry are 17:1-2; 30:15 (cf. vv. 12b, 14b);
and 51:44b-49a.!4
The largest addition in the stratum B material is 39:4-13. Vv. 4-10
add data derived from 2 Kgs 25:4-7, 9-12 (= Jer 52:7-11, 13-16) before
that section was added as an appendix to Jeremiah (chapter 52). At the
same time, it provides new data in vv. 11-13, in which Nebuchad-
rezzar commands Nebuzaradan about Jeremiah.
Of special interest are the deuteronomistic additions in ed. II.
Edition I already contained a deuteronomistic layer, which probably
derived from editor I himself. This deuteronomistic rewriting was
rather extensive, especially in chapters 7, 11, 19, and 21, but it changed
the basic message of the prophet only slightly. The assumption of a
deuteronomustic stratum in Jeremiah (stratum C) is widespread among
scholars, but normally it 1s not realized that this stratum in Jeremiah 1s
composed of two layers. The larger part of the deuteronomistic stratum
is found in ed. I, but editor II added many deuteronomistic phrases (see
section b iv) and also complete sections that abound with deuterono-
mistic phraseology such as 11:7—-8; 29:16-20, and sections of chapter 27
(vv. 7, 13-14a, 17). Editor II may have been one of the last members of
that ill-defined deuteronomistic school,’ or else he simply imitated its
style.
iv. Addition of new details
When analyzing the nature of the added elements, one must pay
attention to the amount of ‘new information’ contained in them. Many of
the added elements somehow derive from the context, but others
contain such data as cannot have derived from the context. It is the
latter that we term new details. These show that editor I] must have
had access to Jeremianic material that had not entered ed. IJ, as
demonstrated by the data mentioned in the previous section with
reference to stratum A. In this section further examples are given,
consisting of a few words only. First, some examples from the prose
sections of the book.
14 10:6-8, 10 (Jacking in the LXX and 4QJer’), on the other hand, may be secondary. The
main topic of the chapter ts a derogation of the idols, and therefore verses 6-8, 10, are
contextually out of place as they extol the God of Israel (if these verses were added
secondarily, they resemble the ‘doxologies’ which have also been added elsewhere in the
Bible). See further Bogaert, “Mécanismes.”
LITERARY HISTORY OF JEREMIAH 369
25:1 The word which came to Jeremiah concerning all the people of
Judah, in the fourth year of king Jehoiakim son of Josiah of Judah,
(which was the first year of Nebuchadnezzar king of Babylon).
Ed. II added a synchronism such as is found often in the historical books.
25:20... all the mixed peoples; (all the kings of the land of Uz.)
25:25... (all the kings of Zimri) and all the kings of Elam.
25:26... all the royal lands which are on the earth. (And last of
all, the king of Sheshach shall drink.)
This last addition remedies the absence of the king of Babylon in the list of kings
and nations that are to drink from the ‘cup of wrath’ (vv. 15-17). He is added here
in ed. II, though in the ‘atbash’ code of secret writing (222 = qwv).1> Likewise, “1
lacking in ed. J, if corrected to -an1, may be an ‘atbash’ form for Elam (o>-y). The
phrase mentioning Uz in v. 20 is lacking in ed. I.
27:19-22. 19For thus said the Lord (of Hosts concerning the columns,
the tank, the stands and) concerning the rest of the vessels (which
remain in this city), 29which (Nebuchadnezzar) the king of Babylon
did not take when he exiled Jeconiah (son of Jehoiakim, king of Judah)
from Jerusalem (to Babylon, with all the nobles of Judah and
Jerusalem—*!thus said the Lord of Hosts, the God of Israel, concerning
the vessels remaining in the House of the Lord, in the royal palace of
Judah and in Jerusalem): 2#They shall be brought to Babylon (and there
they shall remain until I take note of them)—declares the Lord—(and
bring them up and restore them fo this place).
The additions in 27:19-22 stress that the temple vessels which will be carried
off to Babylon will subsequently be returned to Jerusalem. This idea is not
consistent with the spirit of the surrounding verses, which deal with false
prophets and not with the fate of the temple vessels. Even if this were not the case,
it is nevertheless anticlimactic to mention immediately after the threat to the
vessels that ultimately they will be returned to Jerusalem. Nevertheless, from his
point of view, editor II felt that he could not leave the words of Jeremiah without
correction. For the realization of both the threat and the promise, see Dan 5:2-3
and: Ezra 4:7, 11 6:5,
In this section ed. JI speaks of two groups of vessels which were left in
Jerusalem after Jechoniah’s exile and which were to be carried away to Babylon.
Of these vessels, ‘the vessels left in the house of the Lord’ are also mentioned in ed.
I, in 52:17, but the ‘vessels in the house of the king’ are not mentioned there. On the
other hand, in 52:13 Nebuzaradan is said to have burnt ‘the house of the king,’
and as it is known that Nebuchadrezzar took vessels from ‘the house of the Lord’
19 Th the ‘atbash’ system of secret writing, an aleph represents a tav and vice versa, a beth
represents a s(h)in and vice versa, etc. See J.M. Sasson, “Wordplay in the OT,” IDBS, 968-
970, esp. 969.
370 CHAPTER TWENTY-FOUR
before it was burnt, he probably acted similarly with regard to the vessels found
in ‘the house of the king.’
2901 Thus said the Lord (of Hosts, the God of Israel) concerning
Ahab (son of Kolaiah) and Zedekiah (son of Maaseiah,
who prophesy falsely in my name).
The patronymics of the two false prophets are not mentioned elsewhere.
36:22 The king was sitting in the winter house (in the ninth
month).
36:26 ... to arrest Baruch (the scribe) and Jeremiah (the prophet).
36:32 So Jeremiah got another scroll, (and gave it to Baruch
son of Neriah the scribe).
Only in ed. IJ is Baruch explicitly called ‘the scribe.’16
37:17 Zedekiah (the king) sent for him and took him and
questioned him (in his palace) secretly.
It is not stated elsewhere in the context that the meeting took place ‘in his palace.’
38:1 Shephatiah son of Mattan, Gedaliah son of Pashhur,
Jucal son of Shelemiah (and Pashhur son of Malchiah)
heard.
Pashhur the son of Malchiah is known from 21:1, but here he is not mentioned in
ed. I.
38:7 Ebed-melech the Ethiopian, (a eunuch), heard.
Only here is Ebed-melech called ‘eunuch.’
38:12 And [he] said, (Eved-melech the Ethiopian to Jeremiah),
‘Put [them] (the worn cloths and rags under your armpits),
inside the ropes.’
The word for armpits, ny?xx, does not occur in this form elsewhere in the Bible (cf.,
however, *?-xx in Ezek 13:18 and possibly also 72°xx in Ezek 41:8, the word is
further known from rabbinic Hebrew, [-xx, elbow] and Syriac [7>-x, elbow)).
There are similar additions in the poetry sections, but the notion of
‘new details’ in poetry differs from such additions in prose. For the
prose, examples were given of details that provide new factual
information, but little such information is contained in the additions in
poetry. For these, other criteria must be applied. It seems that a ‘new
detail’ in the poetry sections is the addition of any detail that
conceivably could have derived from the prophet himself. It must be
16 Since Baruch is not known from other books of the Bible, it is of interest to mention a
bulla with the inscription 1907 W"7) 72:1772729, ‘belonging to Berechyahu son of Neriyahu
the scribe.’ The title ‘scribe’ is applied to Baruch only in ed. II of Jeremiah (36:26, 32), not in
ed. I. The bulla confirms that Baruch actually bore the title. Although editor II could have
simply inferred this from Jeremiah 36, it is equally possible that he learned this from
authentic traditions about Jeremiah’s life. For the bulla, see N. Avigad, “Baruch the Scribe
and Jerahmeel the King’s Son,” IE] 28 (1978) 52-56.
LITERARY HISTORY OF JEREMIAH 371
remembered that the poetry sections in the book (stratum A), as
opposed to the biographical stratum (B) and the deuteronomistic
stratum (C), are ascribed by all scholars to the prophet himself;
therefore any detail added in stratum A may ultimately go back to an
authentic tradition, but there can be no certainty in these matters. Some
examples follow:
1:18 I make you this day a fortified city, (and an iron pillar), and
bronze walls.
2:2 I accounted to your favor the devotion of your youth, your love
as a bride—how you followed me (in the wilderness, in a land
not sown).
5:15 Lo, lam bringing against you, O House of Israel, a nation from
afar—declares the Lord; (an enduring nation, an ancient nation).
13:17 For if you will not give heed, my inmost self must weep because
of your arrogance (and copiously shed tears), and my eye must
stream with tears.
14:3 They found no water; they returned, their vessels empty (they
are shamed and humiliated, they cover their heads).
31:30 But everyone shall die for his own sins; (every person)
wholfever] eats sour grapes, his teeth shall be blunted.
31:35 Thus said the Lord, who established the sun for light by day,
(the laws of) the moon and stars for light by night.
31:40 (And the entire Valley of Corpses and Ashes), and all the
fields.
46:5 Why (have I seen) they are dismayed, yielding ground.
46:18 As I live—declares (the King), the Lord (of Hosts is his
name).
v. Free rewriting
The recensional differences between the MT and LXX pertain to length
and sequence. But rarely ed. II also rewrote the text
LXX MT
29:25 Thus saith the Lord of Hosts,
the God of Israel:
I did not send you in my Because you sent letters in your
name to own name to all the people in
Jerusalem, to the priest
Zephaniah son of Maaseiah Zephaniah son of Maaseiah
the priest the priest and to the rest of
saying ... the priest saying ...
372 CHAPTER TWENTY-FOUR
35:18
Therefore, thus said the Lord:
Because the sons of Jonadab
the son of Rechab have
obeyed the charge of
their father to do as
their father enjoined them.
36:32 So Baruch got another scroll
and wrote in it.
38:9
You have acted wickedly
in what you have done
to kill this man by starvation.
And to the family of the
Rechabites Jeremiah said:
Thus said the Lord
of Hosts, the God of Israel:
Because you
have
obeyed the charge of Jonadab
your father and have kept all
his commandments and done all
that he enjoined you.
So Jeremiah got another scroll
and gave it to the scribe Baruch
son of Neriah,
and he wrote in it.
O Lord, king,
those men have acted wickedly
in all they did
to the prophet Jeremiah;
they have put him down in the
pit,
to die there of starvation.
Ed. I] probably altered the wording of the previous edition because it blamed
Zedekiah for killing Jeremiah.
44:11-12 Iam going to set my face
to cut off all
the remnant
in Egypt.
b. Exegetical aspects
i. Clarification
] am going to set my face
against you for punishment,
to cut off all
of Judah. I will take
the remnant
of Judah who turned their faces
toward the land to Egypt.
An important aspect of ed. II was the attempt to clarify passages
which apparently struck him as insufficiently clear. The editor read
the book as an exegete and then revised the text, clarifying details in
the context, making explicit what was implicit, and stressing ideas
already found in the book, either in the immediate context or in remote
contexts. In any event, this editor added but few new ideas.
LITERARY HISTORY OF JEREMIAH 373
When inserting the new elements between the existing words of ed. I,
editor II often created syntactic difficulties disturbing the flow of the
Hebrew sentence, as in the following examples.
36:6 And read the scroll (that you wrote at my dictation—the
words of the Lord—) to the people.
41:1 And they ate together there (at Mizpah).
41:2... struck down Gedaliah (son of Ahikam son of Shaphan with
the sword and killed him) whom the king of Babylon had put
in charge of the land.
41:3... and all the Judeans who were with him (with Gedaliah) in
Mizpah.
41:7 He slaughtered them (Ishmael son of Nethaniah) [and threw
their bodies] (in)to a cistern, (he and the men who were with
him).
In other instances, the additions actually contradict the immediate or
remote context:!7
1:3... and in the days of King Jehoiakim son of Josiah of Judah,
until the (end of the) eleventh year of King Zedekiah son of
Josiah of Judah, until Jerusalem went into exile in the fifth
month.
In this addition, the added word non (end), referring to the lower limit of the
prophet’s ministry at the ‘end’ of the eleventh year of Zedekiah, contradicts the
mention of ‘the fifth month’ of that year, also mentioned in 52:5 ff. = 2 Kgs 25:2 ff.
27:1 The title of this chapter contradicts the content of the chapter
itself (see Tov, “Exegetical Notes”*).
27:7 See below, section 3.
28:1 And it occurred (in that year, early in the reign of) Zedekiah
[was] King of Judah in the fourth year in the fifth month
(LXX: And it occurred in the fourth year of Zedekiah King of
Judah in the fifth month).
The sequence of the elements is different in ed. I, but it is clear which elements are
lacking. The added words contradict the chronological indication found in the
continuation of the verse.
29:16—20 A large addition in ed. II.
The greatest contextual difficulty caused by this insertion is that the verse before
the insert (v. 15) has its direct continuation in v. 21 and that vv. 16-20 have no
connection at all with that verse. The added section was inserted in a section that
deals with other matters, although the connecting link can still be recognized. The
verses before and after the addition turn to the exiles, while the added section
17 For further inconsistencies, see 27:19-22 (section a iv); 10:6-8, 10 (p. 363), 28:16
(section b iii); 29:32 (section b iv).
374 CHAPTER TWENTY-FOUR
itself speaks of the people in Jerusalem, even though it is add ressed to the exiles. In
vv. 1-24, the prophet asks the exiles to acquiesce in their situation and to start a
new life. At the same time, in the added section he informs the exiles that the people
left behind in Jerusalem shall perish. The added section and the surrounding
verses thus speak of different matters, but yet are connected: The exiles are urged
to obey the Lord, since their brothers in Jerusalem, who did not obey, will be
punished (v. 20). A further connection between the old and new material is the
mentioning of prophets in both.
ul. Homogenizing additions
The most conspicuous feature of ed. II is its homogenizing filling in of
details that are mentioned elsewhere in the context. Editor II probably
wanted to make the book of Jeremiah as explicit as possible, and he
therefore filled in details that were implicit in ed. I.
a. Personal names
Among the homogenizing additions, the filling in of personal names in
ed. II is the most manifest. Editor II was in the habit of mentioning
personal names in their full form, mainly in the prose sections,
including the name of the father, sometimes also the grandfather, the
title (‘king’ or ‘prophet’), and so on. E.g.,
21:2, etc. (Nebuchadrezzar) the king of Babylon.
28:4, etc. Jeconiah (son of Jehoiakim king of Judah).
28:5, etc. Jeremiah (the prophet) answered Hananiah (the prophet).
36:8, etc. Baruch (son of Neriah) did just as Jeremiah (the prophet)
had instructed him.
40:9, etc. Gedaliah (son of Ahikam son of Shaphan) = 2 Kgs 25:24 MT
and LXX.
52:16, etc. (Nebuzaradan) the chief of the guards = 2 Kgs 25:12 MT and
LXX.
The full evidence for the filling in and addition of names in ed. II is
provided in Janzen, Jeremiah, Appendix A and Min, Minuses and Pluses.
The data must be analyzed not only for the book as a whole but also for
individual literary units. For example, editor I mentioned the full
name or title of the person when he was introduced for the first time in
a given unit, but 1n all or most subsequent references he referred to him
in a shortened form. In this manner editor I follows the practice of
LITERARY HISTORY OF JEREMIAH 375
biblical narrative.}® Ed. II filled in the details of the full formula in
many (sometimes: most or all) occurrences of the name.
A good example of this procedure is ‘Ishmael son of Nethaniah son
of Elishama,’ thus introduced in its full form in ed. I in 41:1, but
shortened in that edition to ‘Ishmael’ in vv. 2, 6, 7, 8, 9 bis, 10, 11, 13, 15,
16, 18. Editor IT left the short name in some verses, but expanded it to
‘Ishmael son of Nethaniah’ in vv. 2, 6, 7, 9, 11, 15, 16, 18. Furthermore,
he added the short name in wv. 3, 10, and 14 and the expanded name in
Vv. IZ.
Likewise, in ed. I Johanan is introduced in chapter 41 as ‘Johanan son
of Kareah’ (41:11), but the next verses refer to him as ‘Johanan’ only
(13, 14, 16). In ed. IIT he is presented in all four verses with the long
form.
Similar filling is visible for Gedaliah, whose name is often
expanded to “Gedaliah son of Ahikam (son of Shaphan)’ (40:6, 7, 9, 11,
14, 16; 41:1, 2, 6, 18), and for Jeconiah, expanded to ‘Jeconiah son of
Jehoiakim king of Judah’ (27:20; 28:4).
When the names in editions I and II are compared, some interesting
details come to light. In ed. I, Jeremiah is usually referred to by his
name only; the fuller description, ‘Jeremiah the prophet,’ occurs but
four times in this edition (42:2; 43:6; 45:1; 51:59). In ed. II this title is
frequently added to Jeremiah’s name (twenty-seven times in all), but
the filling in is systematic only in chapters 28-29. In these two chapters
the prophet is called ‘Jeremiah the prophet’ consistently in ed. II
(eight times) and also once ‘Jeremiah the Anathotite’ (29:27).1?
Edition I of chapter 38 speaks often of ‘the king,’ who according to
the context of chapter 37 can only be Zedekiah. Edition II added this
name in several verses (38:5, 9, 14, 15, 16, 17, 19, 24).
18 See Cross, “Ammonite Oppression,” 111.
In their present form chapters 27-29 manifest some unique features summarized by
N. Sarna, “The Abortive Insurrection in Zedekiah’s Day (Jer. 27-29),” Erlsr 14 (1978) 89-96,
esp. 92. Sarna expresses the view, which had been suggested earlier, that this section ‘must
once have circulated separately’ (ibid.), This suggestion may be somewhat farfetched, but
the evidence (mainly relating to the form of names) displays a scribal tradition that sets the
section off against the remainder of the book. The status of these chapters does not bear on
the issue under investigation, because the special features of these chapters were inserted
before or during the completion of ed. I and they were left intact in ed. If. On the other
hand, the disproportionately large number of additions of the phrase ‘the prophet’ in ed. II
could point to a separate circulation of chapters 27-29 in ed. If. On the other hand, these
chapters provide more occasion than the remainder of the book for the addition of this
phrase as they speak of Jeremiah’s encounter with the false prophets. Accordingly, if
chapters 27-29 circulated separately, this occurred before ed. JI, and the frequent addition
of the phrase ‘the prophet’ occurred as part of ed. I.
376 CHAPTER TWENTY-FOUR
‘The king of Babylon’ is mentioned often in ed. I in this general fashion,
but his identity is made explicit in ed. I] through the addition of
‘Nebuchadrezzar.’ The name Nebuchadrezzar occurs but rarely in ed. I,
mainly in editorial introductions to chapters, so that the prophet
himself probably mentioned this name seldom, if at all.
B. Contextual additions
With his formalistic approach, editor II often did not leave much to
the imagination when he added details in one verse in order to make it
identical with another verse in the immediate context. This homo-
genizing approach, visible especially in the prose sections, is known
from SP and from some of its antecedents among the Qumran
manuscripts.”
135 (I went) and buried it at Perath.
Cf. 13:4 ... and go at once to Perath and bury it there.
13:7b .. and found (the loincloth) [it] ruined.
Cf. 13:7a ... and took the loincloth.
18:6b — Just like clay in the hands of the potter, so are you in my
hands, (O House of Israel).
Cf. 18:6a O House of Israel, can I not deal with you like this potter.
36:15b5 And Baruch read it (to them).
Cf. 36:15a ... and read it to us.
40:6 and stayed (with him) among the people.
Cf. 40:5 ... stay with him among the people.
46:6 In the north by (the river) Euphrates.
Cf. 46:2 ... which was at the river Euphrates near Carchemish.
27:16 The vessels of the House of the Lord shall be brought back
from Babylon (shortly).
Cf. 28:3 (see below).
28:11 So will I break the yoke of (Nebuchadnezzar) the king of
Babylon (in two years).
Cf. 28:3 In two years, I will restore this place all the vessels of the House of the
Lord which Nebuchadnezzar the king of Babylon took from this place and
brought to Babylon.
28:14 _... that they may serve (Nebuchadnezzar) the king of
Babylon—(and serve him they shall! I have even given
the wild beasts to him).
20 Cf. E. Tov, “The Nature and Background of Harmonizations in Biblical MSS,” JSOT 31
(1985) 3-29.
LITERARY HISTORY OF JEREMIAH 377
Cf. 27:6 I have given ... to my servant Nebuchadnezzar the king of Babylon, I even
give him the wild beasts to serve him.
44:1 ... aving in Migdol, Tahpanhes, (and Noph), and in the
land of Pathros.
Cf. 46:14 Declare (in Egypt, proclaim) in Migdol, proclaim in Noph and
Tahpanhes.
Cf. also 2:16 Those, too, in Noph and Tahpanhes will lay bare your head.
lii. Contextual clarifications
Editor II often added clarifying words or phrases explaining a matter
that, in his view, was not clear. For example, the original text of 28:16—
17 (ed. I) was very short and therefore impressive: ‘You shall die this
year; and he died in the seventh month.’ Editor I], however, added the
reason for the verdict and furthermore made it clear that the phrase ‘in
the seventh month’ refers to ‘the same year.’ Ed. II thus reads: ““You
shall die this year (for you have urged disloyalty to the Lord)’; and
(the prophet Hananiah) [he] died (that year), in the seventh month.
Similarly:
27:5 Itis I who made the earth, (and the men and beasts who are on
the earth).
27:8 The nation or kingdom that (does not serve him—Nebuchad-
nezzar king of Babylon—and that) does not put its neck under
the yoke of the king of Babylon.
27:16-17 ... Give no heed to the words of your prophets who prophesy
to you ... for they prophesy falsely to you. (Give them no heed.
Serve the king of Babylon, and live! Otherwise this city shall
become a ruin).
28:3 In two years, I will restore to this place (all) the vessels of the
House of the Lord (which Nebuchadnezzar king of Babylon
took from this place and brought to Babylon).
28:15 And Jeremiah (the prophet) said to Hananiah (the prophet, °
Listen, Hananiah!), The Lord did not send you ...’
29:6 Take wives and beget sons and daughters; and take wives for
your sons, and give your daughters to husbands, (that they
may bear sons and daughters). Multiply (there), do not
decrease.
41:1 In the eleventh month Ishmael ... came to Gedaliah (son of
Ahikam) at Mizpah; and they ate there together (at
Mizpah).
378 CHAPTER TWENTY-FOUR
41:6 Ishmael (son of Nethaniah) went out (from Mizpah) to meet
them, weeping as he walked. (As he met them), he said (to
them), ‘Come to Gedaliah (son of Ahikam).’
41:7 When they came inside the town, (Ishmael son of Nethaniah)
[he] slaughtered them [and threw their bodies] in(to) a cistern,
(he and the men with him).
41:13-14 When all the people held by Ishmael saw Johanan (son of
Kareah) and (all) the army officers with him (they were
glad; all the people whom Ishmael had carried off from
Mizpah turned back), and [they] went over to Johanan (son of
Kareah).
42:9... and said to them: ‘Thus said the Lord, (the God of Israel, to
whom you sent me to present your supplication before him).’
Cf. v. 2 Grant our plea, and pray for us to the Lord your God ...
Among these clarifying additions, one meets many added names (cf.
the amplified names in section ii a [p. 373]):
1:11 The word of the Lord came to me: ‘What do you see,
(Jeremiah)?’ I replied: ‘A branch of an almond tree (I see).’
36:4 So Jeremiah called Baruch son of Neriah; and (Baruch) [he]
wrote down at Jeremiah’s dictation all the words which the
Lord has spoken.
Other names are amplified from pronouns in ed. I:
35:12 The word of the Lord came to Jeremiah (LXX: to me).
37:21... gave instructions to lodge Jeremiah (LXX: him).
52:8 ... and they overtook Zedekiah (LXX: him = 2 Kgs 25:5).
Clarifying amplifications like these are found especially in the
prose sections. A few similar elements are found in the poetry sections,
but these cannot be evaluated well because they may also have derived
from an authentic Jeremianic tradition (see section a 111).
iv. Amplified formulas
Ed. I contained several formulaic expressions, especially at the
beginning and end of prophetic utterances (717° O0N1, declares the Lord;
m7? ox 72, thus said the Lord, etc.). These formulas were often added
in ed. II, which presents a fuller use of these formulas than ed. I. For
example, 717° O81, Which occurs 109 times in both editions I and I, occurs
an additional 65 times in ed. II only. The same applies to 4177 2x 1D.
18:11 And now, say (I pray) to men of Judah and the inhabitants of
Jerusalem (saying: Thus said the Lord:) I am devising
disaster for you.
LITERARY HISTORY OF JEREMIAH 379
27:11 But the nation which puts its neck under the yoke of the king
of Babylon, and serves him, will be left by me on its own soil
(—declares the Lord—) to till it and dwell on it.
29:9 _- For they prophesy falsely to you in my name; I did not send
them (—declares the Lord).
29:11 (For Iam mindful of the plans) I have made concerning you
(declares the Lord)—plans for your welfare.
31:37 (Thus said the Lord) If the heavens above could be measured
The same is true of 19x? (saying), *2x (to me), etc. after verbs of
speaking, and of mxax (Hosts) in various combinations:
1:4 The word of the Lord came to me (saying) ...
Sul (Saying): If a man divorces his wife ...
5:20 Proclaim this to the House of Jacob (saying) ...
1:17 ~~ Arise and speak (to them) ...
13:1. ‘Thus the Lord said (to me): Go buy yourself a loincloth.
17:19 Thus said the Lord (to me): Go and stand in the People’s
Gate.
6:6 For thus said the Lord (of Hosts) ...
ie: Thus said the Lord (of Hosts), God of Israel ...
The latter formula is of particular interest because the full formula occurs thirty-
two times in ed. I, but never in ed. I.
Of special interest are some formulaic deuteronomistic expressions
added in ed. JI (section a iii).*! For example:
7:13 and though I spoke to you (persistently), you would not listen.
35:15 is similar; cf. Bright, “Date,” no. 1.
13:10 This wicked people who refuse to heed my bidding (who
follow the wilfulness of their own hearts).
Cf. Bright, “Date,” no. 6.
19:9 because of the desperate straits to which they will be reduced
by their enemies (who seek their life).
Similarly 34:20; 38:16; cf. Bright, “Date,” no. 11.
21:12 Else my wrath will break forth like fire and burn, with none to
quench it (because of your wicked acts). Cf. Bright, “Date,” no. 14.
23:16 Do not listen to the words of the prophets (who prophesy to
you). Cf. Bright, “Date,” no. 15.
25:3 From the thirteenth year ... 1 have spoken to you persistently,
(but you would not listen). Cf. Bright, “Date,” no. 46.
25:7 But you would not listen to me (declares the Lord; you vexed me
with what your hands made, to your own hurt).
21 For a descriptive list of the vocabulary of the deuteronomistic stratum (the C
stratum) in the book, see Bright, “Date,” 30-35.
380 CHAPTER TWENTY-FOUR
Cf. Bright, “Date,” no. 2.
32:19 to repay every man according to his ways, (and with the
proper fruit of his deeds). Cf. Bright, “Date,” no. 14.
43:5 the entire remnant of Judah who had returned (from all the
countries to which they had been scattered).
Cf. Bright, “Date,” no. 31.
Note further 28:16: ‘You shall die this year (for you have urged
disloyalty [n127 770] to the Lord).’ The phrase 770 727 occurs in ed. II in
29:32 and elsewhere only in Deut 13:6. While the phrase in Deuter-
onomy refers to a prophet who incites to the worship of ‘other Gods’
(Deut 13:3, ‘let us follow other Gods’), the prophets mentioned in
Jeremiah prophesy in the name of the God of Israel.
c. Further characteristics of Edition II
i. Peculiar words and expressions
Ed. II contains some words and expressions that within Jeremiah are
characteristic of that edition only. For example:
1. Nebuchadnezzar is known as ‘God’s servant’ only in ed. II:
25:9 I am going to send for (all) the peoples of the north (declares
the Lord—and for my servant Nebuchadrezzar king of
Babylon), and bring them against this land.
27:6 | herewith deliver (all these) lands to king Nebuchadnezzar of
Babylon, to serve him [1713y?] (MT: *724y my servant); I even give
him the wild beasts.
43:10 Iam sending for (my servant) king Nebuchadrezzar of
Babylon ...
2. The idea that he who serves Nebuchdnezzar will enjoy a long life
is not found in ed. I. The reward of long life is known elsewhere, though
in different contexts, from the sixth commandment, from deuteronomis-
tic phraseology, and from Amos 5:4, 6.29
27:12 WLXX: put your necks and serve the king of Babylon.
MT: put your necks under the yoke of the king of Babylon;
serve him and his people and live.
27:17... give them no heed. Serve the king of Babylon and live ...
3. ]w71, ‘fatness, fat ashes,’ occurs within Jeremiah only in ed. I:
31:14 I will give the priests their fill (of fatness).
22 See Tov, “Exegetical Notes,”*and Bogaert, “Mécanismes.”
23 For the idea of longevity in the Bible and the ancient Near East, see Weinfeld,
Deuteronomy, 257, 308-9; A. Malamat, Israel in Biblical Times (Hebr.; Jerusalem 1983) 295-306.
LITERARY HISTORY OF JEREMIAH 381
31:40 (And the entire Valley of Corpses and Ashes []w7]) and all
the fields ...
4. The phrase p.N7 ov occurs only in ed. II in 34:19 (52:25 may be
disregarded, because that chapter has been transferred from 2 Kings 24-
25 as an appendix to Jeremiah [note Jer 51:64)).
34:19 and (all) the people (of the land) ([paxq] ov [9>))).
5. Within the Bible the term 72n(7) 725, ‘the king’s commanders,’
occurs only in ed. II of Jeremiah:
39:13... and Nergal-sarezer the Rab-mag, and all the commanders
of the king of Babylon (9213 7?n 727).
41:1 Ismael son of Nethaniah son of Elishama, who was of royal
descent (and one of the king’s commanders [92n7 ~2])*4 came
with ten men ...
The component 35 used in this phrase is probably parallel to the other terms
compounded with 25 that are mentioned in 39:13. Note that the phrase is absent
from 2 Kgs 25:25, which is otherwise identical with Jer 41:1.
ll. Resumptive repetition (‘Wiederaufnahme’)
When editor II inserted several new elements, and when the idea of the
original text might be lost because of the insertion, he occasionally
repeated the lead phrase or its approximate contents. A similar
technique termed Wiederaufnahme or ‘resumptive repetition’ has been
recognized in the redaction of biblical books.*° Note, for example, Jer
27:21 (See section a iv): After the long additions in vv. 19 and 20, ed. I]
felt the need to repeat the introductory formula of the prophecy as well
as the object of the prophecy:
19For thus said the Lord (of Hosts ... 4Thus said the Lord of
Hosts, the God of Israel, concerning the vessels ...)
Another example of resumptive repetition:
28:3-4 3In two years, I will restore to this place (all) the vessels of
the House of the Lord (which Nebuchadnezzar king of Babylon took
from this place and brought to Babylon). 4And Jeconiah (son of
Jehoiakim king of Judah) and (all) the Judean exiles (who went to
Babylon, I will bring back to this place—declares the Lord).
24 This reconstruction is based on the assumption that dd yévoug Tov Baothéwe (‘from
the offspring of the king’) represents 7319n7 yn», ‘from the offspring of the kingship’ (cf.
the LXX of 26[33]:1).
25 Cf. C. Kuhl, “Die ‘Wiederaufnahme’—Ein literarkritisches Prinzip?” ZAW 64 (1952)
1-11; LL. Seehgmann, “Hebraische Erzahlung und biblische Geschichtsschreibung,” TLZ 18
(1962) 305-325, esp. 314-324; Talmon, “Textual Study,” 395, n. 174.
382 CHAPTER TWENTY-FOUR
In this verse, the repetition of ‘I will bring back to this place’ (cf. v. 3a) was
needed after the long additions in vv. 3b-4a.
41:2-3 *Then Ishmael (son of Nethaniah) and the ten men who
(were) with him arose and struck down Gedaliah (son of Ahikam son of
Shaphan with the sword and killed him), whom the king of Babylon
had put in charge of the land, and all the Judeans who were with him
(with Gedaliah) in Mizpah and the Chaldeans who were stationed
there (the soldiers, Ishmael struck down).
Since the object of 139, ‘they struck down’ (v. 2) was greatly expanded, ed. II
repeated the subject and the verb at the end of the sentence in v. 3.
41:10 Ishmael carried off (all the rest of the people who were in
Mizpah), the daughters of the king, all the people left in Mizpah over
whom (Nebuzaradan) the chief of the guards, had appointed
Gedaliah son of Ahikam (and Ishmael son of Nethaniah carried them
off) and set out to cross over to the Ammonites.
After the object was expanded, ed. II repeated the verb and its subject.
c. The tendencies of edition II
Some of the tendencies of ed. IT are visible throughout the book,
especially in chapters 10, 25, 27-29, 33, 39, 52:26
i. Many of the additions emphasize the guilt of the nation, deriving
from its frequent rebellion against the Lord, and for which it will be
punished. E.g., 11:7-8; 29:16—20; 32:30.
ii. The centrality of God, referred to everywhere in the Bible, but
even more so in ed. II. Ed. I of chapter 10 derides the idols of the
heathen, while vv. 6-8, 10 of the LXX and 4QJer® extol the Lord of
Israel. Ed. II stresses the central role of God in the history of Israel.
Everything happens according to his command, and even Nebuchad-
nezzar is ‘his servant’ (see Tov, “Exegetical Notes”* on Jer 27:6). See
further the task of Babylon in God’s plan for the world (25:14).
ili. Ed. II added many actualizing details. The punishment of
Babylon after seventy years is mentioned in 25:14 and 27:7. Ed. I
26 See P.-M. Bogaert, “Urtext, texte court et relecture: Jérémie xxxiii 14-26 TM et ses
préparations,” in: J.A. Emerton, (ed.), Congress Volume, Leuven 1989 (VTSup 43; Leiden
1991) 236-247; id., “Le livre de Jérémie en perspective: Les deux rédactions antiques selon
les travaux en cours,” RB 101 (1994) 363-406; Goldman, Prophétie; H.-J. Stipp, Das
masoretische und alexandrinische Sondergut des Jeremiabuches—Textgeschichtlicher Rang,
Eigenarten, Triebkrafte (OBO 136; Freiburg /G6ttingen 1994); L. Stulman, The Prose Sermons of
the Book of Jeremiah—A Redescription of the Correspondences with Deuteronomistic Literature in
the Light of Recent Text-critical Research (SBL Dissertation Series 83; Atlanta, GA 1986); R.D.
Wells, “Indications of Late Reinterpretation of the Jeremianic Tradition from the LXX of Jer
21 1—23 8,” ZAW 96 (1984) 405-420.
LITERARY HISTORY OF JEREMIAH 383
mentions the exile of the temple vessels, but ed. II also mentions their
return to Jerusalem (27:21—22). The postexilic date of ed. Il is evident in
various late additions (see section 3 below). See further Goldman,
Prophétie, 65-105 on Jer 29:16—20.
iv. Ed. II seems to be interested in priestly subjects. The temple
vessels are mentioned briefly in ed. I in chapters 27 (e.g., v. 19) and 52,
but very extensively in ed. II. See further 33:14—26.
v. Ed. II is interested to show that the prophet’s utterances are
fulfilled: 25:14; 27:7, 19-22.
3. The postexilic date of edition II
The exilic date of some passages in ed. | is evident (e.g., 8:3; 9:15; 16:13;
16:15; 30:10-11). Other passages may be postexilic (esp. 25:11; 29:10). At
the same time, there are indications of the postexilic date of ed. II:
25:14 For they too shall be enslaved by many nations and great
kings; and I will requite them according to their acts and
according to their conduct.
27:7 All nations shall serve him, his son and his grandson—until
the turn of his own land comes, when many nations and great
kings shall subjugate him.
Both the above verses, added in ed. II, foreshadow the fall of the Babylonian
empire and were probably added as vaticinia ex eventu.2/
27:19-22 See section a iv.*8
29:6 Take wives and beget sons and daughters; and take wives for
your sons, and give your daughters to husbands, (that they
may bear sons and daughters). Multiply (there), do not
decrease.
Through the addition the prophet refers to an exile that would last for three
generations, but it is not certain that this was indeed intended by the addition.
Bogaert and Lust*? refer to the appearance in ed. II of details that
show that some time had lapsed between the composition of editions I
and II.
The above description shows that ed. II inserted many minor and
major changes in Jeremiah, both in the editing and rearranging of the
27 See Tov, “Exegetical Notes.”*
28 The addition mentions, among other things, the return from the exile and thus
betrays its late origin. For a similar addition, see 29:14, and on a different level, 40:12. On
these additions, see the discussion of J. Lust, “’Gathering and Return’ in Jeremiah and
Ezekiel,” in Bogaert, Le lrure de Jérémie, 119-142.
29 Bogaert, “Mécanismes,” 236-37, referring to the postexilic emphasis on the priests in
ed. II in 33:14-26; 27:16, 19. For Lust, see the previous note.
384 CHAPTER TWENTY-FOUR
material and in its clarification. Editor II rearranged the text, added
headings to prophecies, repeated sections, added new material, and
inserted some changes in content. In the clarification of the text, he
added homogenizing details (especially in personal names) and
clarified details in the context. The sources for his additions are the
context, his imagination, but also genuine Jeremianic material which
somehow found its way into ed. IJ. The additions were inserted neatly
in between the elements of ed. I, and sometimes the editor added so
many elements that after them he had to repeat the last words that
had preceded them (resumptive repetition). In other cases, however,
the secondary character of the insertion is still visible from its
formulation or content. All these changes were inserted during the final
stage of the growth of the book, thus shedding light not only on textual
criticism, but also on literary criticism.
CHAPTER TWENTY-FIVE
THE GROWTH OF THE BOOK OF JOSHUA IN LIGHT OF THE
EVIDENCE OF THE SEPTUAGINT
The LXX of Joshua reflects many pluses, minuses, and differences
which, when retroverted into Hebrew, present a book different from
that contained in MT. The divergences are not as comprehensive as
those in the book of Jeremiah (see Tov, “Jeremiah’”*, but their content is
often very important. Also 4QJosh? differs considerably from MT (see n.
35):
The LXX of Joshua is important not only for the textual criticism of
the Hebrew book, but also for its literary criticism. Many scholars have
noticed the deviations of the LXX from MT, but most are not ready to
recognize the contribution of the LXX to the literary criticism of Joshua.
They continue to regard the LXX as a textual witness only. The
approach of these scholars is eclectic: some deviations of the LXX are
ascribed to the translator, while others are recognized as reflecting
possible original readings, especially when they comply with the
scholar’s views on the original form of the book. Such was the
approach of Dillmann, Noth,! and Wright—Boling in their commen-
taries.2 Wright and Boling recognized the majority of the minuses of
the LXX, but they did their best to ascribe as many as possible to the
presumed inclination of the translator to haplography and homoto-
teleuton.
This, however, is not the only approach to the LXX of Joshua. During
the past seventy years several important studies have been written by
scholars who recognized its significance for literary criticism. The first
to do so was Holmes who in an eighty-page study displayed much
understanding and intuition in textual matters. His monograph
provides a valuable introduction to this topic as well as a running
LA Dillmann, Numeri, Deuteronomtum und Josua (2d ed.; Leipzig 1886); M. Noth, Das Buch
Josua (3d ed.; HAT; Tubingen 1971).
2RG. Boling and G.E. Wright, Joshua (AB; Garden City, NY 1982).
S. Holmes, The Hebrew and Greek Texts of Joshua (Cambridge 1914).
386 CHAPTER TWENTY-FIVE
commentary to the text. In the same breath we should mention Cooke,4
who applied Holmes’ views in a running commentary to the Hebrew
text.
The special contribution of the LXX was studied also by Orlinsky in
a methodological study focusing on the minuses of the LXX.° In a series
of studies on the LXX, Auld® usually accepted the evidence of the LXX
as original, an opinion thus formulated in one of his articles: ‘And
where MT and LXX differ in the book of Joshua, the latter is generally
to be preferred.’”
Of two innovative studies by A. Rofé, the first one deals with the
extensive Greek plus at the end of the book, after 24:33, long since
recognized as reflecting a Hebrew addition,® and whose text can be
retroverted easily into Hebrew. Rofé described the ancient elements in
this plus of the LXX, elements so unusual from a contextual point of
view that they were rejected by the editor of MT. Indeed, this plus
mentions, inter ala, the flint knives buried in Joshua’s grave, a
tradition which looks like the preserving of reliquiae, possibly rejected
by a later generation. One also notes that the last words of this plus
quote a phrase of Judg 3:12, viz., ‘and God delivered them into the
hands of Eglon, king of Moab and he ruled them eighteen years.’ The
plus in Joshua also contains other phrases from Judges, viz., from the
ideological introduction to that book: ‘and the Israelites worshipped
the Ashtarot (and Ashtoret) and the gods of the nations around them’
(cf. 2:11,13). On the basis of these data, Rofé turned to the audacious
view, which had been suggested earlier, that the plus in the LXX
reflects an earlier and more original stage in the development of
Joshua—Judges in which the two books were combined and in which the
first chapters of Judges were lacking. The LXX thus reflects an ancient
tradition, in which the end of Joshua was followed by Judges 3. It then
4 G.A. Cooke, The Book of Joshua (CB; Cambridge 1918). For fine insights on the
contribution of the LXX to the literary criticism of Joshua, see also C. Steuernagel, Lehrbuch
der Emlettung in das Alte Testament (Tiibingen 1912) 276-287.
2 HLM. Orlinsky, “The Hebrew Vorlage of the Septuagint of the Book of Joshua,” VTSup
(1969) 187-195.
CAG. Auld, Sfudtes in Joshua, Text and Literary Relations (unpubl. diss.; Univ. of
Edinburgh, 1976); Moses and the Land—Tetrateuch-Pentateuch-Hexateuch in a Generation since
1938 (Edinburgh 1980); “The Levitical Cities: Texts and History,” ZAW 91 (1979) 194~206;
“Cities of Refuge in Israelite Tradition,” JSOT 10 (1978) 26-40; “Textual and Literary
Studies in the Book of Joshua,” ZAW 90 (1978) 412-417.
7 “Judges I and History: A Reconsideration,” VT 25 (1975) 264.
8 “The End of the Book of Joshua according to the Septuagint,” Henoch 4 (1982) 17-35 =
Shnaton 2 (1977) 217-227 (Hebrew).
THE GROWTH OF THE BOOK OF JOSHUA 387
becomes necessary to show, as does Rofé, that the first two and a half
chapters of Judges contain secondary material.’
A second article by Rofé deals with the cities of refuge in chapter
20.19 It was recognized long ago that the large minus of the LXX in 20:4-—
6 presents a special text. That chapter speaks about the cities of refuge,
and it would be interesting to know on which of the Pentateuchal law
codes the chapter is based. It so happens that the long text of MT
reflects the terminology and quotations from both the Priestly Code in
Num 35:9-34 and Deut 19:1-13. In the LXX, on the other hand, the
quotation from Deuteronomy is lacking in 20:4-6. It was therefore
suggested by Rofé, as well as by other scholars, that an earlier stage of
Joshua, such as reflected in the LXX, referred only to the Priestly
formulation of the law of the cities of refuge. According to these
scholars, the later edition of the book, as reflected in MT, added the
terminology of, and a quote from, Deuteronomy.
Several important pericopes in the book have been studied by L.
Mazor, who depicted the literary development of Joshua on the basis of
these newly gained insights.!!
When returning to the general questions relating to the LXX of
Joshua, we note that the major feature characterizing the Greek
translation is its many minuses, and secondly, its pluses. This situation
reminds one of other books in the LXX, especially Jeremiah.
The minuses of the LXX of Joshua are not as numerous as those in 1
Samuel 17-18 (see Tov, “Samuel’*) and in Jeremiah (see Tov, “Jere-
miah”*). In Joshua the LXX lacks not more than 4-5%, a proportion
similar to that in Ezekiel (See Tov, “Ezekiel’”*). The elements lacking
in the LXX are sometimes very significant. At the same time, the LXX
contains important pluses, such as after 16:10; 19:47; 21:42; 24:30; 24:33.
? 1:1-2:5 contain deviating traditions about the conquest, parallel to the book of Joshua
and to Judges 17-18; Judg 2:6-3:6 are likewise secondary as they contain the ideological
deuteronomistic introduction to the book, and 3:7-11 present a story about a ‘judge’
Othniel and an otherwise unknown and probably imaginary king Qushan Rish‘atayim (note
the etymology of the name), about whom nothing tangible is said, a story which may very
well have been introduced to exemplify the course of events in the book of Judges.
10 “Historico-Literary Criticism Illustrated by Joshua 20,” Isac Leo Seeligmann Volume (eds.
A. Rofé and Y. Zakovitch; Jerusalem 1983) 137-150 = “Joshua 20—Historico-Literary
Criticism Illustrated,” in: Tigay, Models, 131-147.
! L. Mazor, The Septuagint Translation of the Book of Joshua—Its Contribution to the
Understanding of the Textual Transmission of the Book and Its Literary and Ideological
Development, Hebrew University, Jerusalem, 1994 (Heb. with Eng. summ.); id., “The Origin
and Evolution of the Curse upon the Rebuilder of Jericho—A Contribution of Textual
Criticism to Biblical Histonography,” Textus 14 (1988) 1-26.
388 CHAPTER TWENTY-FIVE
These pluses are written in a manifestly Hebraistic diction and they
are retroverted rather easily into Hebrew.
In addition to these pluses and minuses, the LXX reflects a major
difference in the sequence of events. The section dealing with the
erection of the altar in Shechem appears in the LXX not in 8:30-35 as in
MT, but after 9:2 (see Tov, “Sequence”*).
Since external evidence is lacking regarding the evaluation of the
special features of the LXX of Joshua, we are left with internal
evidence as our only guide. For this purpose the translator’s word-
choices must be examined as well as the word order and his consistency
in the translation. The problem of the evaluation of the LXX is thus
reduced to that of its translation technique. A modest beginning has
been made by Hollenberg and Mazor.!4
The translation character of the LXX cannot be determined through
an analysis of the pluses and minuses, for these are problematic
themselves. We thus turn to an overall analysis of the translation
technique, especially to the choice of translation equivalents and their
consistency. That translation technique is known to range from
‘relatively free’ to ‘relatively literal,’ yet remains sufficiently close to
the Hebrew to establish the translator’s faithfulness to his source.!4
It is further noteworthy that the pluses of the LXX are Hebraistic in
diction (see Tov, TCU, 83-84), and they can be retroverted easily into
Hebrew.!4 Thus, according to the logic presented earlier, it is not
feasible for one translator to have faithfully rendered the text and at
the same time omit significant elements. Moreover, no principle can be
recognized for a supposed shortening by the translator.
12 J. Hollenberg, Der Charakter der alexandrinischen Ubersetzung des Buches Josua und ihr
textkritischen Werth (Berlin 1876); L. Mazor (see n. 11 above).
3 In the wake of Tov-Wright, “Literalness,’* the criteria described in the article were
applied to Joshua as well. Thus it was found that in 58.4% of the instances, -2 was
rendered by €v, in 53.2% *D was rendered by 6TL, and in 83.1% the pronominal suffixes of
the third person were rendered by forms of avtég and éautdés. These data should be
compared with the data for the other books of the LXX. In addition to these absolute
figures there is also comparative material with regard to the addition of prepositions in the
LXX and the relation between kal and the post-position particles such as 6€ and ydp. On
the basis of these data, Joshua should be ranked as relatively free to relatively literal. At the
same time, the freedom of the translator is often predictable, so that the reconstruction of
its Hebrew base text is often easier than shown by mere Statistics. See further the
conclusion of Sollamo, Semiprepositions, 285, who includes Joshua in the second of four
groups (relatively free), together with Leviticus, Genesis, Numbers, and Deuteronomy.
Further arguments in favor of retroverting elements of the Hebrew parent text of the LXX
are prover’ by L. Mazor’s dissertation mentioned in n. 11.
1 See, e.g., the retroversions in BHK, BHS, and in Rofé, “Historico-Literary Criticism”
(see n. 10).
THE GROWTH OF THE BOOK OF JOSHUA 389
In the wake of the above considerations, we now phrase the following
working hypothesis: the LXX of Joshua reflects a text which differed in
several major details from MT, both in its minuses, pluses, and
substitutions. This working hypothesis is strengthened by an internal
analysis of the evidence itself.
The Hebrew base of the LXX is now compared with MT, using
literary criteria. This study focuses on the minuses of the LXX which
are much more frequent than its pluses. For the sake of convenience we
speak of the ‘short’ text of the LXX, even though the LXX is not much
shorter than MT. The working hypothesis suggested here is that a
short text like the LXX was expanded to a long text like MT. The use of
the word ‘like’ enables us to account also for pluses of the LXX.!°
On the basis of these arguments it is suggested that the elements
hitherto described as minuses of the LXX actually are pluses of MT.
This is exemplified mainly from the first half of the book.
We first turn to a category of readings showing the secondary nature
of the pluses of MT. All the pluses of MT (minuses of the LXX) are
written in parenthesis.
1. Additions of MT whose secondary nature 1s evident from the context
4:10 The priests who bore the ark remained standing in the middle of
the Jordan until all the instructions that the Lord had ordered Joshua to
convey to the people had been carried out (just as Moses had ordered
Joshua).
According to the short formulation of the LXX, Joshua’s actions closely followed
the command of God, while the plus stresses that the command was by Moses. The
juxtaposition of these two commands in MT is not impossible—after all, God
commanded Moses to command Joshua (Deut 3:28)—but it is awkward. It is
therefore likely that two different remarks have been combined in MT. Possibly
the plus in MT derived from Deut 3:28 or from v. 12 in the context. The
juxtaposition of two similar elements recurs in Josh 1:13-14 where, according to
the short text of the LXX, God gave the land to the two-and-a-half tribes, while
according to the long text of MT it was given by Moses.
5:11-12 They ate of the produce of the land (on the day after the
Passover offering), nxn and parched grain, on that very day. '*The
manna ceased (on the day after) when they ate of the produce of the
land.
15 Also the ‘short’ text of Jeremiah is at times a little longer than the ‘long’ text of MT.
390 CHAPTER TWENTY-FIVE
According to the previous verse (5:10), the Israelites ‘offered the Passover
sacrifice on the fourteenth day of the month in the steppes of Jericho.’ That is,
according to the common text of v. 10 and the short text of v. 11, the Israelites ate
the Pesach on the fourteenth of Nissan in the evening, and on the same occasion
they ate nxn and parched grain. On the other hand, according to the addition of
MT in v. 11, the Israelites ate their ny on the next day, the fifteenth of Nissan.
The addition of MT apparently adapted the original text to the regulations of
Lev 23:5 which prescribes the Pesach on the evening of the fourteenth and myn on
the next day. The present text of MT thus contains both elements, so that tension
exists between the phrases ‘on that very day’ and the plus of MT ‘on the day after
the Passover offering.’ The editor of MT paid no attention to this tension when
adding the latter phrase.
8:11-13 All the warriors who were with him advanced near the city
(and encamped to the north of Ai with a valley between them and Ai.
12nd he took about five thousand men and he stationed them in an
ambush between Bethel and Ai, west of the city. '7And the people set
the whole camp north of the city) and its rear west of the city (and
Joshua went that night into the middle of the valley).
The plus of MT speaks of an encampment to the north of Ai, but according to v. 9
the people are to the west of that city, between Bethel and Ai. Likewise, according
to the plus in v. 12, ‘He (scil. Joshua) took about five thousand men and stationed
them in an ambush between Bethel and Ai, west of the city (Q: Ai).” MT thus
contains at least two versions of the ambush. According to the version in vv. 3-9,
common to the LXX and MT, the ambush occurred west of the city and some 30,000
men were involved. On the other hand, according to the plus of MT in wv. 11b, 12,
13a, the ambush took place north of the city and some 5,000 men participated in it.
That second version, lacking in the LXX, was added in MT, and contradicts the
first account.
2. Additions in MT whose secondary nature is evident from their
formulation
Most of the examples refer to exegetical expansions of MT.
1:7... to observe faithfully (all the teaching) that (which) my ser-
vant Moses enjoined upon you. Do not deviate from it to the right or to
the left. |
1:15 Then you may return to the land of your possession (and you may
possess it) which Moses the servant of the Lord assigned to you.!®
5:2... and circumcise the Israelites again (for the second time).
16 The relative ‘which’ returns immediately to its antecedent ‘possession.’ The inter-
vening sentence has been added.
THE GROWTH OF THE BOOK OF JOSHUA 391
5:6 ... because they had not obeyed the Lord who had sworn (the
Lord) them.
10:35 They captured it on the same day and put it to the sword and
all the people that were in it he proscribed (on the same day).
10:39 ... just as they had done to Hebron, he did to Debir and its king
(and as he had done to Libnah and its king)—an afterthought.
13:21 ... and the entire kingdom of Sihon, the king of the Amorites
(who reigned in Heshbon), whom Moses defeated.!”
The main section of examples, paragraphs 3-9, illustrate various
types of exegesis.
3. Small elucidations
3:12 (Now) select twelve men ...
3:9-10 And Joshua said to the Israelites, “Come closer and listen to
the words of the Lord your God.’ !9(And Joshua said) ‘By this you shall
know ...’.
5:1 ... and no spirit was (left) in them because of the Israelites.
7:2 And Joshua sent men ... east of Bethel (and he said to the
Israelites) as follows ...
1:2 Prepare to cross <the> (this) Jordan : (417) 77777 Mx Tay OP
1:4 from the wilderness and <the> (this) Lebanon:(417) 7113277) Da7nnn
10:23 ... <the> (these) five kings: (7?xn) O°3207 Nwnn NRX
1:2 ... into the land which I give to them (to the Israelites).
1:11 ... prepare (for yourself) provision.
4:2 ... select (for yourself) from among the people twelve men.
4. Harmonizing additions!®
2:4 The woman took the (two) men—cf. v. 1.
2:5 The men left, and I don’t know where <they> (the men) went—<f.
Vida:
2:9 I know that the Lord has given the country to you, because dread
of you has fallen upon us (and all the inhabitants of the land are
quaking before you)—cf. v. 24: the Lord has delivered the whole land
into our power; in fact, all the inhabitants of the land are quaking
before us.
17 The first of the two relative clauses is secondary.
18 For a definition and analysis of this category, see E. Tov, “The Nature and Back-
ground of Harmonizations in Biblical MSS,” JSOT 31 (1985) 3-29.
392 CHAPTER TWENTY-FIVE
2:21 She said: ‘Let it be as you say.’ She sent them (and they left and
she tied the crimson cord to the window)— cf. v. 18: you tie this length of
crimson cord to the window
6:17 Only Rahab the harlot is to be spared, and all who are with
her in the house (for she hid the messengers we sent)—<f. v. 25: for she hid
the messengers that Joshua sent to spy out Jericho.
7:18 and Achan son of Zerah son of Zabdi (of the tribe of Judah) was
indicated—cf. v. 17.
5. Contextual additions
1:13-14 The Lord your God is granting you a haven; he is assigning
this territory to you. !4Let your wives, children and livestock remain in
the land which <he> assigned to you (Moses across the Jordan).
2:1 Joshua son of Nun (secretly) sent two spies...
2:15 She let them down (by a rope) through the window.
6:1 Now Jericho was shut up tight (because of the Israelites).
10:11 The Lord hurled huge stones on them from the sky, all the way
to Azekah (and they perished); more perished from the hailstones
than were killed by the Israelite weapons.
10:26 And Joshua smote them (afterwards and he killed them) and
impaled them ...
6. Emphasis!
1:7 But you must be (very) strong and resolute.
2:5 Go after them (quickly), for you can overtake them.
2:17 We will be released from this oath (which you have made us
take).
3:1 <they> (he and all the Israelites) came to the Jordan.
6:10 Do not let your voices be heard (and do not let a sound issue from
your lips).
7. Substantial additions
2:4 And she said: ‘It is true, the men came to me (but I did not know
where they were from).’
According to the short text, Rahab affirms that people came to visit her; she did not
ask the men any questions nor did she justify the fact of their visit. The addition of
MT, however, gives Rahab an opportunity to justify her deeds.
19 These examples have been singled out because of their implication for the analysis.
THE GROWTH OF THE BOOK OF JOSHUA 393
2:15 She let them down (by a rope) through the window (for her
house was at the outer side of the city wall and she lived in the actual
wall [NJPSV]).
The second part of the verse explains a detail which is necessary in the context.
The remark explains how the explorers left the city quietly without being noticed:
Rahab’s house was located in such a place that the explorers could easily leave
the city.2° As a consequence, v. 15b, which through the ages has inspired numerous
exegetical conjectures, actually belongs to a secondary layer of the text, not found
in the LXX.41
The same conclusion is reached through an internal analysis of MT of another
chapter, viz., chapter 6. With the fall of the walls of Jericho Rahab’s house also
should have collapsed if indeed her house was joined to the walls. Nevertheless,
the implication of 6:22 is that Rahab’s house was still intact after the walls fell:
‘To the two men who spied the land Joshua said: “Go into the house of the harlot
and bring out from there the woman and all whom she has as you have sworn to
her.”’ Consequently, it looks as if Rahab’s house was still intact in chapter 6 and
was therefore not joined to the wall as mentioned in the plus of MT in 2:15b.
10:13 And the sun stood still and the moon halted, while a nation
wreaked judgment on its foes (as is written in the Book of Jashar). Thus
the sun halted in mid heaven ...
The plus of MT (‘as is written in the 1w77 750’) disturbs the context, but this fact
alone is insufficient proof of its late origin. The secondary nature of these elements
is further indicated by their absence from the LXX. Also Solomon’s benediction,
spoken at the time of the dedication of the temple (1 Kgs 8:12-13, occurring in the
LXX after v. 53), is reportedly contained in the qw7n 150, this time according to the
LXX only (€v BiBALw TAS wWSrs, reflecting wn 1902 deriving from Wn 1902). The
absence of these words in MT underlines the textual instability of the ascription
of biblical compositions to the qw7n 150. The only place in which the reference to
the Ww 190 occurs in all of the textual sources is 2 Sam 1:18 (David’s lament over
Saul and Jonathan).
20 The note itself is tautological.
! The short text in 2:15 ‘She let them down through the window’ implies that Rahab’s
house was near the wall, because otherwise it is unclear why she would have let the spies
down from the window. The explanation for this short text reflected in the MT of v. 15b
presents one possible view of the short text, but not the only one. Possibly Rahab’s house
was near the wall and when the two men were let down through the rear side of the
house they could flee more easily than through the front door. Another possibility is that by
letting the men leave by the window, possibly of a second floor, the people in the house
would not know that they had left.
394 CHAPTER TWENTY-FIVE
8. Theological corrections
Some additions present theological corrections.
4:5 Walk up to (the ark of) the Lord (your God).
6:7 ... and the vanguard marched in front of (the ark of ) the Lord.
6:13 ... and the seven priests bearing the seven ram’s horns marched
in front of (the ark of) the Lord.
7:6 And he fell on his face to the ground in front of (the ark of) the
Lord.
In several places the original (short) text mentioned actions happening ‘in front of
the Lord,’ referring to the actual presence of the Lord with the ark of the covenant.
The origina] idea, which implies direct contact with God, has been toned down in
MT by the addition of the words ‘the ark of.’
9. Influence of Deuteronomy
Some of the additions of MT reveal influence from Deuteronomy, both in
direct quotes and in its characteristic deuteronomistic formulations. A
direct quote is found in 20:4-6 (cf. Deut 19:46). Note also the following
deuteronomistic formulations among the pluses of MT:
1:1 After the death of Moses (the servant of the Lord) ...
Cf. also 1:15; 12:6; 22:4 (cf. Driver, Introduction, 116).
1:7 .... to observe faithfully according to (all the teaching) which
Moses my servant enjoined upon you. Cf. Weinfeld, Deuteronomy, 336.
1:11 The land which the Lord your God gives to you (as a possession).
Cf. ibid., 314.
4:10 The priests who bore the ark remained standing in the middle of
the Jordan until all the instructions that the Lord had ordered Joshua to
convey to the people had been carried out (just as Moses had ordered
Joshua).
24:17 For it was the Lord our God who brought us and our fathers up
from the land of Egypt (the house of bondage, and who wrought those
wondrous signs before our very eyes).
We now turn to an analysis of the overall nature of the LXX of Joshua.
By way of a working hypothesis it was suggested to consider the
elements lacking in the LXX as pluses in MT. The first two categories of
examples showed that many of them are contextually secondary in MT.
Furthermore, the other groups of examples reflect exegetical additions
to the text. These, too, are by definition secondary.
22 For ovay ma cf. ibid., 326-327, and for a7ndw1 minx, ibid., 330.
THE GROWTH OF THE BOOK OF JOSHUA 395
The additions of MT have to be analyzed in light of other phenomena
in the LXX. The LXX of Joshua also contains some added verses, which
may have been omitted from MT, and furthermore it reflects inversions
of some verses and also of one section, viz., at the end of chapter 8 of
MT. The major pluses of the LXX are in 16:10 (cf. 1 Kgs 9:16 [5:14 LXx]);
19:47-48; 21:42 (cf. 19:49-50; 5:2-3); 24:30 (cf. 5:2-3); and 24:33 (cf. Judg
2:6, 11-14; 3:12). In one case the possible reason for the omission of these
verses from MT may be indicated, viz., the mentioning of the flint
knives in the pluses after 21:42 and 24:33.
The omissions and additions of the Vorlage of the LXX make the MT
and LXX into two parallel editions differing with regard to their
content. As a rule, the additions can easily be accounted for. Thus in the
geographical lists, the LXX contains an added verse in 16:10 about the
conquest of Gezer by Pharaoh—this verse is almost identical to 1 Kgs
9:16 (5:14 LXX). In Josh 19:47-48, elements have been added in the
Vorlage of the LXX describing the migration of the Danites to the
North. The added elements run parallel to Judg 1:34-35. In yet another
section, the plus after 21:42 summarizes the division of the land
described in the earlier chapters. The LXX thus contains two such
summaries, for 19:49-50 contain the same summary. The following
development may be reconstructed: the original summary is found in
19:49-50, both in the MT and LXX. When, at a later stage of the
development of the book, chapters 20 and 21 were added, dealing with
the cities of refuge (20) and the Levitical cities (21), an additional
summary was appended at the end of chapter 21 in the Hebrew base of
the LXX. This section more or less repeats 19:49-50, and it has not been
added in the edition of MT.
The data adduced here lead to the view that the MT and LXX do not
reflect textual differences, but rather two different editions of the book.
The differences between the two editions are not great with regard to
the book as a whole, but in individual contexts they are. An analysis of
the minuses of the LXX leads to the conclusion that the edition of MT
expanded the shorter one reflected in the LXX. According to this
description, the two editions are related genetically. The pluses of the
LXX do not contradict this assumption, but they show that the edition
of MT expanded an earlier edition very similar to the Vorlage of the
LXxX.
Further research on the geographical details may bring us closer to a
solution. The issue of the date of these editions should not be confused
with that of their textual attestation, which is much later than the
editions themselves. Thus the short edition of the LXX is first attested
396 CHAPTER TWENTY-FIVE
in the LXX translation prepared in the late third or early second
century BCE, and that of MT is attested around the turn of the era.
Little can be said about the nature of the expansions of the edition of
MT. Its main characteristic is the addition of exegetical remarks and
traditions. Among these, a few deuteronomistic formulations stand out.
It is remarkable that as late as the third or second century BCE,
when the LXX was produced, the short Hebrew edition was still
available to the translators.23 The only possible explanation for this
situation 1s that even after the short edition was altered to the edition
of MT, it was not destroyed, but continued to exist. In other words, when
the short edition was expanded, more than one copy of that edition
already existed. Not all of these were discarded, so that its impact
continued to be felt afterwards.
It is probably no coincidence that the short text was preserved in
Egypt, far away from Palestine. Even when a new edition was accepted,
it did not oust the previous one in Egypt, where it remained in use for
centuries. This description serves as a modest reformulation of the local
texts theory, especially connected in recent years with the names of
Albright and Cross.*4 Even if one does not accept the validity of that
theory as a whole, at least the aspect described here is plausible.
23 A third tradition in Joshua is reflected in 4QJosh*, which has a shorter text than MT,
similar to that of the LXX, in frgs. 9 ii and 13-16 (Josh 8:3-14, 18?). Furthermore, the section
which in MT narrates the building of an altar after several actions connected with the
conquest (8:30-35), is located at an earlier place in the story in 4QJosh®, before 5:1,
immediately after the crossing of the Jordan, and probably also in Josephus, Antiquities,
V:16-19. According to Rofé and Ulrich this sequence of events in 4QJosh®, which probably
reflects the original story, shows that the Qumran text constituted a third independent
text of Joshua, alongside MT and the LXX: A. Rofé, “The Editing of the Book of Joshua in
the Light of 4QJosh2,” in: Brooke—Garcia Martinez, New Qumran Texts, 73-80; E. Ulrich,
“4QJoshua® and Joshua’s First Altar in the Promised Land,” ibid., 89-104.
4 WF. Albright, “New Light on Early Recensions of the Hebrew Bible,” BASOR 140
(1955) 27-33; Cross, “Evolution” (earlier formulations by Cross are mentioned there).
CHAPTER TWENTY-SIX
RECENSIONAL DIFFERENCES BETWEEN THE MASORETIC TEXT
AND THE SEPTUAGINT OF EZEKIEL
There exist many differences in details between the MT and LXX of
Ezekiel which are usually ascribed to copyists of the Hebrew text or
the Greek translator. In light of this assumption, Cooke remarked: ‘In
the Hebrew Bible perhaps no book, except 1 and 2 Samuel, has suffered
more injury to its text than Ezekiel.’! This is actually an overstatement
as many of the differences between the MT and LXX were created at the
time of the literary growth of the book, and therefore should not be
ascribed to textual factors. In this regard the LXX of Ezekiel resembles
other biblical books and pericopes whose relevance to literary criticism
has been discussed in recent years (see Tov, TCHB, chapter 7). This
study focuses on a few select issues in Ezekiel which highlight the
recensional differences between the two main preserved texts.
1. Ezek 7:3-9
The first nine verses of chapter 7 are represented according to the RSV
in such a way that the parallel structure of vv. 34 and 8-9 is stressed:
The word of the Lord came to me:
2. ‘And you, O son of man, thus says the Lord
God to the Jand of Israel: An end! The end
has come upon the four corners of the land.
3. Now the end is upon you, 8. Now [ will soon pour out my wrath upon you
and J will let loose my anger upon you, and spend my anger against you,
and will judge according to your ways; and judge you according to your ways;
and I will punish you and |] will punish you
for all your abominations. for all your abominations
1G A. Cooke, Ezekiel (ICC; Edinburgh 1936) XL.
2 For special studies in addition to the commentaries, see J.A. Bewer, “On the Text of
Ezekiel 7:5-14,” JBL 45 (1926) 226-231; J. Goettsberger, “Ezek. 7:1-16 textkritisch und
exegetisch untersucht,” BZ 22 (1934) 195-223.
398 CHAPTER TWENTY-SIX
3. And my eye will not spare you, 9. And my eye will not spare
nor will I have pity; nor will I have pity;
but I will punish you for your ways, I will punish you according to your ways,
while your abominations are in your midst. while your abominations are in your midst.
Then you will know that] am the Lord. Then you will know that I am the Lord,
who smite
5. Thus says the Lord God: Disaster after
disaster! Behold it comes.
6. Anend has come, the end has come; it has
awakened against you. Behold it comes
7. Your doom has come to you, O inhabitant
of the land; the time has come, the day
is near, a day of tumult, and not of
joyful shouting upon the mountains.
It is not easy to summarize the content of vv. 1-11 in MT. These verses
contain a prophecy of doom, or a series of doom prophecies. The first
nine verses constitute a string of ‘alarms of doom,’ as they are called by
Greenberg.’ The sequence of ideas in MT is difficult, as vv. 3-4 and 8-9
are almost identical, and there does not seem to be a literary reason for
the repetition. The content of the LXX is equally difficult, and
probably even more so. For the LXX has the identical sections, vv. 3-4
and 8-9, next to each other; in fact, vv. 8-9 of MT precede vv. 34 in the
LXX. This juxtaposition is difficult, if not impossible. From a contextual
point of view it is not logical to assume with Cooke, Ezekiel, 76 that
the translator, or his Vorlage brought the two identical sections
together. Rather, a textual mishap such as a doublet must be presup-
posed here. As in other instances in the LXX (see Tov, “Sequence”*), the
different sequence of the LXX may point to a late insertion of a section.
Probably one of the two parts of the doublet was added in MT in one
place and in the LXX in another. At first the added section was placed
in the margin and from there it reached two different places in the text.
This assumption solves two problems, that of the unusual repetition in
all texts and of the difference in sequence between the MT and LXxX. If
this suggestion is correct, we are faced with a textual-literary problem.
Basically a doublet is a textual issue, but the duplicates were probably
not created by scribes. They could have derived from different literary
editions of the prophecy in the formative period of the book. There
may even be additional cases of such duplication during this period,
that is, during the stage of the literary development of the book, but in
3M. Greenberg, Ezektel 1-20 (AB; New York 1983) 157.
RECENSIONAL DIFFERENCES IN EZEKIEL 399
those cases the duplication occurs only in MT and there is no textual
witness reading only one of the two parts of the doublet. Such instances
have been listed by Kraetzschmar.*
The assumption of a different text underlying the LXX in chapter 7 is
further supported by other major differences between the two texts in
this section as well as elsewhere in the chapter. The LXX lacks several
phrases in this section in vv. 4, 5, 6, 7, 10, and 11 as well as elsewhere in
the chapter (vv. 13, 14, 19, 24), which undoubtedly were also lacking in
the Vorlage. The combination of the different sequence in chapter 7 and
the minuses of the LXX makes the assumption of a recensionally
different text likely. Zimmerli, in the introduction to his commentary”
calls this a ‘badly disturbed’ text; however he also says: ’... the
question arises whether already at the initial phase, before the
redactional composition of the prophet’s words into a book, the
individual sections were extant in differing qualities of preservation.’
2. Minuses and pluses
With two exceptions,® the issue of the ‘short’ text of the LXX of Ezekiel
has not been studied. The problem is not as obvious as in the case of
Jeremiah and the story of David and Goliath, which in some chapters
lacks as much as 45% of the text of MT. In Ezekiel the combined minuses
of the LXX do not amount to more than 4-5% of the text. Since similar
problems have been treated extensively for the short texts of Jeremiah,
the story of David and Goliath, and Joshua,’ the main argument is
repeated briefly. The issue at stake is whether the quantitative
differences between the MT and LXX of Ezekiel were created by the
translator or were already extant in his Hebrew Vorlage. Since there is
no external evidence relating to the quantitative differences, we turn to
translation technique. Translators were to a great extent internally
consistent in their approach, and therefore, when encountering large or
recurrent differences between the MT and LXX we turn to that issue. If a
translator rendered his Vorlage with relative literalness, he would not
4 Ezechiel (HKAT ; Gottingen 1900) xiii. E.g., 1:1,3a//1:23b; 1:13//1:14; 4:10,11,16,17/ /
4:9,12-15; 9:5//9:7. Kraetzschmar goes one step further when claiming that the book of
Ezekiel as a whole consists of two parallel recensions combined in the period prior to our
earliest textual evidence.
> W. Zimmerli, Ezekiel 1 (Philadelphia 1969) 75.
6 J. Lust, “The Use of Textual Witnesses for the Establishment of the Text--The Shorter
and Longer Texts of Ezekiel, An Example: Ez 7,” in: Lust, Ezekiel, 7-20; P.-M. Bogaert, “Les
deux rédactions conservées (LXX et TM) d’Ezéchiel 7,” in: Lust, Ezekiel, 21-47.
7 See Tov, “Samuel,”* “Jeremiah,”* and “Joshua.’’*
A400 CHAPTER TWENTY-SIX
have omitted so many words and even parts of verses. If, on the other
hand, a translation is recognized as free, it is not surprising that
phrases were added and omitted. The evaluation of these minuses in
the LXX of Ezekiel is thus reduced to that of its translation technique.
In that case we are on safe ground, for the translation of this book is
relatively literal and consistent, as has been determined by studies on
word order, consistency in the choice of lexical equivalents, as well as
in the choice of certain equivalents.®
Evidence from translation technique thus supports the view that the
short text of the LXX reflects a short Hebrew Vorlage which is not
known from manuscripts, including 11QEzek.? But a content analysis of
the short readings does, in our view, support the aforementioned view.
In most instances the short text of the LXX reflects a more original text
from a contextual point of view, and the long text of MT a secondary one.
Since we rule out the possibility that the otherwise literal translator
of Ezekiel was involved in shortening, MT should be considered
expanded. Rather than taking the LXX as a short text, we should thus
take MT as an expanded text, as is exemplified below by several
categories of amplification of MT. The amplifications of MT represent
an added layer of contextual exegesis, clarification and slight editing.
These instances should not be explained as scribal mistakes, although
other minuses of the LXX of Ezekiel may reflect occasional mistakes.!!
The English translation follows, as much as possible, that of Greenberg,
Ezekiel (see n. 4). Pluses of MT (that is, minuses of the LXX) are written
within parenthesis.
8 See G. Marquis, The Translation Technique Reflected in LX X-Ezekiel, unpubl. M.A. thesis,
Hebrew University, Jerusalem, 1982; idem, “Word Order as a Criterion for the Evaluation
of Translation Technique in the LXX and the Evaluation of Word-Order Variants as
Exemplified in LXX-Ezekiel,” Textus 13 (1986) 59-84. Sollamo, Semtprepositions included
Ezekiel in a group of ‘slavishly’ translated books. My own study of certain translation
equivalents in the book of Ezekiel found the evidence inconclusive, recognizing relatively
inconsistent types of renderings together with very consistent ones. See Tov—Wright,
“Literalness.”* Likewise, the study of running texts supports the assumption of a very
literal translation. C.H. Cormill, Das Buch des Propheten Ezechiel (Leipzig 1886) 97 ff. and J.W.
Wevers, Ezekiel (New Century Bible; London 1969) 30 mention the faithfulness of this
translation.
9 See W.H. Brownlee, “The Scroll of Ezekiel from the Eleventh Qumran Cave,” RevQ 4
(1963) 11-28 and the final publication of this text by E.D. Herbert in DJD XXIII (Oxford
1998). The text of the scroll is too fragmentary for extensive evaluations. Usually the scroll
goes with MT rather than the LXX, but in 5:13 the fragment lacks the same two words as
the LXX (nomi *ninin) [based on count of letters]).
Cin very few cases (mainly 1:4, 22; 24:14; 27:34; 37:1) the LXX reflects a text longer than
MT. Our formulation should thus always be cautious, referring to the ‘generally’ shorter
text of the LXX and the ‘generally’ longer text of MT.
1] See 1:23, 25-26; 6:4, 5; 7:11; 8:18-9:1; 10:9, 21; 16:6; 23:42; 24:3, 6; 30:3.
RECENSIONAL DIFFERENCES IN EZEKIEL 401
a. Contextually secondary elements
2:1—2 Get on your feet and I shall speak to you (2) Spirit entered me
(as he had spoken to me) and got me on my feet.
The addition of the words ‘as he had spoken to me’ (*?8 127 Wx3) disturbs the
natural connection between the words ‘Spirit entered me’ and ‘and got me on my
feet.’ WwxD probably means ‘as’ in the comparative sense and not in the temporal
sense since that use is not found elsewhere in Ezekiel, }2 ~OX 727 WWKND means ‘as he
had spoken to me’ as elsewhere in the Bible (Gen 7:9; 18:33; 20:13, etc.). The plus
of MT was probably secondary, as God did not tell the prophet in so many words
that Spirit would enter him; these words disturb the context and are imprecise.
2:3 1 am sending you to the sons of Israel, (to the nations) the
rebellious ones who have rebelled against me ...
Although 071) is used of Israel also elsewhere in Ezekiel, it is not impossible that
its addition here creates a certain softening of the next word, ‘the rebellious ones,’
since it may include other peoples. The formulation of the addition itself, 707
o-n with the article added to the second word only may betray its late origin,
since this is the linguistic custom in rabbinic Hebrew. !9
2:4 (The sons—brazen-faced and tough-hearted—I am sending you to
them) and you shall say to them ...
This long addition disturbs the direct connection between the end of the preceding
verse and nvnx1 in v. 4. According to v. 3, the prophet is sent to the sons of Israel
and the immediate continuation of that thought is at the end of v. 4: ‘and you shall
say to them.’ The idea of the plus in v. 4 derives from the immediate context as well
as from 3:7; Exod 33:3, 5; 34:9; Deut 9:6, 13. Furthermore, the plus ‘I am sending
you to them’ is unusual since this thought has already been expressed in the
previous verse.
3:1 He said to me: ‘Man, (whatever you find, eat). Eat this scroll and
go, speak to the house of Israel.’
Commands as well as their implementations are frequently repeated in the Bible,
and even more often, phrased identically. Cf., e.g., later in the chapter (vv. 22-23):
‘The hand of YHWH came upon me there, and he said to me: “Get up and go out to
the plain and there | shall speak to you” (23) So I got up and went out to the plain
and there was the Majesty of YHWH.’ Likewise, our verse is construed in MT by
way of a parallel command and its implementation. However, the parallelism is
secondary in the context. For the prophet does not find anything; he is given
12 Comill, Ezechtel, ad loc. (see n. 11 above).
Constructions of this type occur also rarely in biblical Hebrew. See P. Jouon,
Grammaire de |'Hébreu biblique (Rome 1947) 429. In rabbinic Hebrew, however, they are
much more frequent. See M.H. Segal, A Grammar of Mishnatc Hebrew (Oxford 1927) 182-85.
402 CHAPTER TWENTY-SIX
something, for which cf. T (amnn7). The addition of MT may well be based on Jer
15:16 OF) JIT NEN.
5:16 When I let loose against them my (deadly) arrows of famine,
which are for destroying (which I will let loose to destroy you; and
more famine will I add upon you), I shall break (for you) the staff of
bread.
The first part of the long plus repeats the first stich. The second part, referring to
famine, rephrases the idea of the ‘staff of bread’ mentioned later in the verse.
6:12 He who is far off shall fall by the plague, he who is nearby
shall fall by the sword (and he who remains) and the besieged shall
die by famine.
The original pattern of the verse mentioned an opposition between jn and 3Pn
to which 110 and WNw3n was added. For the first two groups single threats are
mentioned, while the third threat is connected with two groups: 11810 and Kwai.
Probably also the third threat was originally uttered for one group only, and in
that case Iwi is secondary.
8:7-8 He brought me to the entrance of the court (and there I saw a
hole in the wall). (8) He said to me, ‘Man, burrow (through the wall).’
I burrowed (through the wall), and there was an entrance.
The additions disturb the context. There was no hole in the wall when the prophet
came there, since only afterwards was the hole created. Commentators discuss this
difficulty and some of them distinguish between a hole (1m) as seen first by the
prophet and an opening (mn) created by the prophet who enlarged the hole. This
opposition between nn and nnp is, however, not natural. Furthermore, the size of
the hole is not important; its sole purpose was to allow the prophet a viewing
point. If we are attentive to the differences between the short text of the LXX and
the longer one of MT, we recognize the following stages in the development of the
text: according to the short text of the LXX, the prophet is told to dig, and a nnd is
created. According to the long text, a hole existed from the outset; the prophet was
told to dig and as a result a nnd was created. No mention is made in the text of the
‘enlarging’ of the original hole, although according to MT the prophet actually
enlarges that hole. This tension within MT betrays the secondary nature of the
plus.
10:2 He said to the man clothed in linen (and he said): ...
The repetition of Wx") is awkward.
10:7 He (the cherub) reached his hand (from among the cherubs) to
the fire that was among the cherubs.
This verse creates many problems of understanding, and beyond these the
repetition of the phrase ‘from among the cherubs’ is difficult. What is the meaning
of the statement that the cherub reached out his hand from among the cherubs to the
fire that was among the cherubs? The short text of the LXX is preferable.
RECENSIONAL DIFFERENCES IN EZEKIEL 403
10:22 As for the shape of their faces, they were the faces I saw at
the Chebar canal (their appearances) and themselves.
The combination of 0x1 (07°x10) is linguistically difficult.
23:32 You shall drink your sister’s cup which is deep and large (you
shall be laughed at and held in derision) for it contains much.
The plus causes a separation between the two parts of the sentence. In the original
formulation these contained the simile of the cup, but the plus has a different
imagery.
b. Addition of parallel words and phrases
2:3 ... who have rebelled against me, they and their fathers (have
transgressed against me) to this very day.
The verbs 779 and vw® are parallel in the sense of political rebellion, with yw
having the added meaning of religious transgression. The words *2 1wo in MT
constitute a plus since Omax1 nan belong together (cf. Jer 9:15).
5:13 My anger will be spent and (I will assuage) my fury against
them.
MT’s plus *n1117) is probably lacking also in 11QEzek (based on a count of
letters). The word is added on the basis of identical phrases in 16:42; 21:22(17);
24:13.
5:14 I will turn you into a ruin (and a reproach) among the nations
that are around you.
Note the parallel status of 729n (ruin) and 757n (reproach). For a similar
situation, see Jer 49:13 (29n?) 759n? nw? *D where the last word is lacking in the
LXX.
5:15 ... when I execute judgments against you (in anger and in fury)
and by chastisements of fury. Cf. Deut 29:27.
6:6 ... so that your altars shall be ruined (and shall be desolate).
6:6 Your incense braziers shall be hewn down (and what you have
made wiped out).
7:20 And they made images of their abominable (loathsome) things.
8:11 Each with his censer, and the smoke (cloud) of incense
ascending.
any, translated here as ‘smoke,’ is a hapax legomenon. ]1y (cloud) was probably
added as an explanatory gloss (from Lev. 16:13?).
16:22 And with all (your abominations and) your harlotry you did
not remember the time of your youth.
23:33 A cup of horror (and desolation) is the cup of your sister Samaria.
CE.33:°28° 35°53:
24:14 I will not go back, I will not spare, (I will not repent).
404 CHAPTER TWENTY-SIX
24:16 You shall not mourn or weep (nor shall your tears run down).
c. Exegetical additions
1:12 Each went straight ahead; to the place where the spirit would
go (there) they went.
5:16 When I let loose against them my (deadly) arrows of famine.
6:11 Clap your hand and stamp your foot and say ‘Ah!’ over all the
(evil) abominations of the house of Israel.
8:1 As I was sitting in the house with the elders of Judah seated
before me, the hand of the Lord YHWH fell upon me (there).
8:9 Go in and see the (evil) abominations that they are committing
here.
8:10 I went in and saw that there were all sorts of (figures of creeping
things and beasts), detestations and all the idols of the house of Israel.
Cf. Deut 4:17-18.
8:11 ... and seventy men of the elders of the house of Israel were
standing before them—with Jaazaniah son of Shaphan (standing)
among them.
8:12 ... for they say, ‘YHWH does not see (us).’
The short text has a meaning of its own, fully supported by Ezek 9:9 where the
same phrase occurs.
8:13 He said to me, “You will yet see (again) great abominations.’
8:15 You will yet see (again) abominations even greater than these.
8:16 ... whose backs were to YHWH’s temple and whose faces were
turned east and they were prostrating themselves (eastward) to the
sun.
10:1 I saw that on the expanse above the heads of the cherubs was
something as of sapphire; (what appeared to be) the figure of a throne
(was visible) above them. Cf. 1:26.
23:28 For thus says the Lord YHWH, ‘Behold I will deliver you into
the hands of those whom you hate (into the hands) of those from whom
you turned in disgust.
28:25 When I gather (the house of) Israel from the peoples...
29:3 | am against you, Pharaoh (king of Egypt). Cf. v. 2.
d. Contextual clarification
The pluses of this group explain difficulties in the context. Otherwise
the examples resemble those of the preceding group.
RECENSIONAL DIFFERENCES IN EZEKIEL 405
1:3 ... in the land of the Chaldeans by the Chebar canal and the
hand of YHWH came upon him (there).
1:21 ... and when these rose off the ground [they] (the wheels) rose
alongside them.
5:13 My anger will be spent and (I will assuage) my fury against
them (and I will satisfy myself).
6:9 ... their heart which whores (which turned) away from me and
at their eyes that whored after their idols.
6:10 And they shall realize that 1 YHWH (not for nothing) spoke (to
do this evil to them).
The short version was very concise: "N27 7171" "1X 7D 17). The long version was
probably influenced by 14:23: 72 "Mwy wx 7D DX wy Oi x? *D ONYT.
6:13 ... on every high hill (on every mountaintop) and under every
luxuriant tree (and under every leafy oak), the place where they
offered soothing savors to all their idols. Cf. Jer 2:20; 3:6.
8:12 Do you see, man, what the elders of the house of Israel are doing
(in the dark), each in his image-chambers.
9:4 And he (YHWH) said to him.
9:8 As they were killing (I alone was left) I threw myself down on
my face.
10:16 When the cherubs raised their wings to rise off the ground the
wheels did not change their position (beside them).
23:45 But righteous men shall pass judgment on them with the
sentence of adulteresses and with the sentence of (women that shed)
blood, because they are adulteresses and blood is upon their hands.
28:23 (For I will send upon her) pestilence and blood (is) in her
streets.
29:14 I shall bring them back to the land of Pathros, the land of
their origin, and they shall be (there) a lowly kingdom.
e. Harmonizing pluses
1:8 ... as for the faces (and the wings) of the four of them. Cf. 1:11 (and
their faces) and their wings were separated.
Feld See: 1:8:
1:15 As I looked (at the living creatures), I saw one wheel on the
ground. Cf. 1:14, 15b, etc.
1:16 And the appearance of the wheels (and their design) were like
chrysolite. Cf. 1:16b (and their appearance) and their design.
1:16b Cf. 1:16a.
1:17 (When those went) these went on their four sides. Cf. 1:19 When
the creatures went the wheels went beside them.
406 CHAPTER TWENTY-SIX
6:9 They will loathe themselves (for the evil things they did) for
all their abominations. Cf. 20:43 you will loathe yourselves for all the evil
things that you did.
7:19 They shall fling their silver into the streets; their gold shall be
as an unclean thing (their silver and shall be powerless to save them on
the day of YHWH’s rage). Cf. Zech 1:18 their silver and gold will both be
powerless to save them on the day of YHWH’s rage.
8:2 I looked, and there was a figure (having the appearance) of fire;
from (the appearance of) his loins down was fire, and from his loins up
was something (with a brilliant appearance) like ?nwn. Cf. 1:27 From the
appearance of his loins upward I saw the like of ?nwn (having something with the
appearance of fire surrounding it); and from the appearance of his loins
downward ... 1:4 out of the fire appeared something that looked like Snwn.
8:16 There were about twenty (five) men. Cf. 11:1 There at the entrance of
the gate were twenty-five men.
10:18 The majesty of YHWH came forth from (the threshold of) the
house. Cf. 10:4 The majesty of YHWH rose off the cherub onto the threshold of the
house.
13:7 Surely it is idle visions you have uttered and false divination
you have spoken, (who say, ‘declares YHWH,’ when I never spoke). Cf.
13:6 They utter idle visions and false divination who say, ‘declares YHWH’ when
YHWH did not send them.
f. Emphasis
3:14 A wind lifted me and took me, and I went (bitter), my spirit
raging.
8:6 Do you see what they are doing, the great abominations (that
the house of Israel) (they) are doing here.
22:27 Her princes in the midst of her are like wolves tearing the
prey, shedding blood, (destroying lives) to get dishonest gain.
25:6 Because you have clapped your hands and stamped your feet
and rejoiced (with all the malice).
29:12 And I will make the land of Egypt a desolation in the midst of
desolated countries; and her cities shall be among the cities that are
laid waste, (as a desolation) for forty years.
37:25 They shall dwell in the land where your fathers dwelt that |
gave to my servant Jacob, and they shall live in it (and their children
and their children’s children, for ever).
RECENSIONAL DIFFERENCES IN EZEKIEL 407
g. New material
The examples to be discussed in this section are more problematic than
those preceding, not so much because of their content, but because of the
idea underlying this section. The pluses described so far showed that
the editor of MT took liberty in adding his comments to the text. These
comments somehow derived from the context, so that the editor’s
actions are understandable. However, in the items discussed here he
added new material, and a similar claim may be made with regard to
some of the examples classified above as ‘mere’ exegesis. It is not
impossible that the editor of MT drew on oral or written traditions
which included some elements not incorporated in the earlier, short
edition. Similar developments have been recognized in Jeremiah and
Joshua (see n. 7).
1:22 There was a shape over the heads of the creature, of an expanse
that looked like (dreadful) ice.
1:24 I heard the sound of their wings as they went like the sound of
the deep sea like the voice of the Almighty (a sound of tumult like the
sound of an army).
1:27 I saw the like of ?nwn (having something with the appearance
of fire surrounding it).
7:16 Those of them who escape shall haunt the mountains (like
doves of the valleys), all of them moaning [LXX: I shall kill], each in
his iniquity.
9:6 They started with the (old) men who were before the house.
10:12 (All their flesh) and their backs and their hands and their
wings and the wheels were filled with eyes.
13:5 You have not gone up into the breaches and made a fence about
the house of Israel so that they can stand (in battle) on the day of
YHWH.
16:13 You were very very beautiful (fit to be a queen). Cf. 15:4.
20:28 They saw every high hill and every leafy tree, and there they
made their sacrifices (there they placed their vexatious offerings).
h. Deuteronomistic formulations
The suggestion that deuteronomistic formulations have been added in
the last formative stage of the growth of the book, that is the stage of
MT, is intriguing, and has been pointed out so far for Joshua and
Jeremiah (see n. 7). The material collected for Ezekiel is scanty and
problematic. In fact, scholars do not even agree that the MT of Ezekiel
408 CHAPTER TWENTY-SIX
contains deuteronomistic language, let alone that the LXX is relevant to
this issue. However, when we follow the claim of those—like
Herrmann!4+—who believe that MT contains some deuteronomistic
phrases, one notes that some of these are lacking in the LXX:
8:17 They fill the land with lawlessness (and vexed me again
["10°VD77 12w%]); indeed, they reach the vinebranch to their noses.
8:18 For my part, I will react in fury, my eye shall not spare nor will
I have pity, (they shall cry loudly for me, but I will not listen to them
[OMX vow X77) DY CNND NAP).
See further the wording of the added layer of 5:15 (cf. Deut 29:27) in
section b and of 8:10 (cf. Deut 4:17-18) in section c.
j. Formulaic language
11:5 ... and he said to me: (Say), this said YHWH.
20:26 ... so that I might desolate them (so that they might know
that Iam YHWH).
23:38-39 they have defiled my sanctuary (on the same day) and
profaned my sabbaths. (39) For when they had slaughtered their
children in sacrifice to their idols, (on the same day) they came into my
sanctuary.
27:2 (You), son of man, raise a lamentation over Tyre.
30:6 (Thus says YHWH): ‘Those who support Egypt shall fall.’
36:7 Therefore (thus says the Lord YHWH) ...
36:23 ... and the nations will known that Iam YHWH (says the Lord
YHWRH).
3. Chapter 36 in P. Chester Beatty 967
P. Chester Beatty 967 of the LXX lacks vv. 23c-38 in chapter 36, a
section which proclaims the promise of a new heart. The main idea of
this section is expressed in vv. 25-27: ‘I will sprinkle clean water upon
you, and you shall be clean from all your uncleannesses; and from all
your idols I will cleanse you. 2A new heart I will give you, and a new
spirit I will put within you; and I will take out of your flesh the heart
of stone and give you a heart of flesh. 77And I will put my spirit within
you, and cause you to walk in my statutes and be careful to observe my
ordinances’ (RSV).
14.5. Herrmann, Die prophetischen Heilserwartungen im Alten Testament (BWANT V, 5;
1965) 264-65.
RECENSIONAL DIFFERENCES IN EZEKIEL 409
How are we to evaluate the lack of a complete section (36:23c—38) in P.
Chester Beatty dating from the second or early third century CE? This
papyrus reflects a pre-hexaplaric text, that is, it lacks the revisions
towards MT inserted in the Hexapla later in the third century. From
the outset we are thus favorably disposed towards the evidence of P.
Chester Beatty in chapter 36. Moreover, LaW'' also lacks exactly these
verses, and thus lends further support to the thesis that the OG lacked
this section. Indeed, also elsewhere La preserved many early readings,
especially in the historical books.!° Furthermore, J. Lust developed a
theory according to which that section was lacking not only in the OG,
but also in its Hebrew source.!®
This is a far-reaching assumption, because it implies that the OG
reflects an early stage in the development of the Hebrew book of
Ezekiel. Indeed, Lust argued for the lateness of this section in MT. Thus,
only in this section does one find *D1x (also found in Jeremiah);
elsewhere, Ezekiel uses “1x. Likewise, 79W37 *nyv) MONIT 7712 In 36:36
reminds one of Jeremiah’s language and not of that of Ezekiel (e.g., Jer
1:10; 18:7; 24:6). Further reminiscences of Jeremiah are the phrase
O>"MAX? -NN) WR prXa(v. 28), for which cf. Jer 7:7; 16:15, etc.; and
a>?7un) (v. 31), typical of the C stratum of Jeremiah. In vv. 25, 29 the
plural form of 7Xnv is used, while elsewhere in the book the singular is
employed (22:15; 24:13; 36:17; 39:24). In conclusion, according to Lust,
36:23c-38 is based on ideas in the surrounding chapters, on 11:19 and
reminds one of Jeremiah’s vocabulary.
This presumed late intrusion in the Hebrew book of Ezekiel was also
added subsequently in the Greek textual tradition. Thus the pre-
hexaplaric witnesses such as P. Chester Beatty reflect the short
original text, while the expanded text is found in the post-hexaplaric
Greek sources.!” If the evidence of P. Chester Beatty and La can indeed
be trusted, the OG lacked a section which 1s secondary in the Hebrew
15 See the studies listed in Tov, Lucian,”” n. 32.
16) Lust, “De Samenhang van Ez. 36-40,” Tydschrift voor Theologte 20 (1980) 26-39; id.,
“Ezekiel 36-40 in the Oldest Greek Manuscript,” CBQ 43 (1981) 517-33. See also P.-M.
Bogaert, “Le témoignage de la Vetus Latina dans |’étude de la tradition des Septante:
Ezéchiel et Daniel dans le papyrus 967,” Btb 59 (1978) 384-95.
17 Y¢ this view is correct, ideally one would find differences in vocabulary between the
presumably hexaplaric vocabulary (karge-Th?) of the added section and the prehexaplaric
vocabulary of the surrounding chapters. Such an attempt was made by H.St.J. Thackeray,
“The Greek Translators of Ezekiel,” JTS 4 (1903) 398-411 who was guided merely by his
intuition, since in his time the textual evidence on chapter 36 was not yet known. Among
other things, Thackeray pointed to the transliteration of *17N in this section (36:33,37) as
opposed to its translation by KUpLtog elsewhere in Ezekiel.
410 CHAPTER TWENTY-SIX
text of Ezekiel and this information is essential for our understanding of
the literary growth of the book.
4. Conclusion
If the above analysis is correct, the MT and LXX of Ezekiel reflect two
different redactional stages of the book,!® even if the quantitative
differences are not as extreme as in the case of Jeremiah. In the past
these minuses have often been described as glosses or interpolations in
MT, but such a description is irrelevant.!? Redactional differences have
been found in quantitative matters (additions of MT recognized through
muinuses of the LXX), different arrangement and the long text of chapter
7 in MT, and a substantial addition in the MT of chapter 36 if the
textual evidence has been interpreted correctly. Of special interest are
the aforementioned links between Jeremiah and the layer which
possibly was added in MT. Further links may be recognized in
additional places as well.29 The additional layer of MT added
exegetical remarks and harmonizing details, explained contextual
difficulties, and in one instance changed the sequence of the text. As we
are confronted here with different stages in the literary development
of the book (preserved in textual witnesses), no reading should be
preferred textually to that of another, as is customary among most
scholars.?}
18 Thus in a seminal way already A. Merx on the basis of very limited textual evidence:
“Der Werth der Septuaginta fur die Textkritik des Alten Testaments, am Ezechiel
aufgezeichnet,” JPrTh 9 (1883) 65-77, esp. 75; Wevers, Ezekiel, 30; Lust and Bogaert, as
mentioned in n. 6 above.
In fact, such glosses or interpolations have been detected more in Ezekiel than in any
other book of the Bible, and the model of Ezekiel negatively influenced the analysis of the
other books. See Tov, “Glosses,”* II, 4.
0 For the phrase "> 70 in 6:9, cf. Jer 32:40 (not elsewhere); for the added references to
idolatry in 6:13, cf. Jer 2:20, 3:6; for the plus of MT in 3:1, cf. Jer 15:16; and for the plus of
MT in 5:14, cf. Jer 49:13, where a similar element has been added in MT.
21 This argumentation is developed in detail in TCHB, 347-349.
CHAPTER TWENTY-SEVEN
SOME SEQUENCE DIFFERENCES BETWEEN THE MASORETIC
TEXT AND THE SEPTUAGINT AND THEIR RAMIFICATIONS FOR
LITERARY CRITICISM
The MT and LXX often differ regarding the sequence of one or more
verses or chapters. In such instances, a unit of MT may appear in the
LXX in a different place, either in the direct vicinity or in an adjacent
chapter. The neutral term ‘sequence difference’ is used in this context,
and not ‘transpositions,’ since the Jatter takes either MT or the LXX as
point of departure. A partial list of these sequence differences has been
provided by Swete, Introduction, 231-242. The full evidence is included
in the CATSS database (see Tov, “Computerized Database”*).
The present study is concerned with the evaluation of some major
sequence differences, with special attention to their importance for
literary criticism. From the outset we would expect only the most free
and paraphrastic translator to insert in the translation such drastic
changes as major changes of sequence. However, even paraphrastic
translators such as those of Isaiah, Daniel, and Esther did not change
the sequence of the text. Consequently, sequence differences in other
books should not be ascribed to the translators, as in the case of
Numbers, Joshua, 1 Samuel, 1 Kings, Jeremiah, and Ezekiel to be
discussed below. The examples adduced below are intended to establish
a connection between some sequence differences and literary develop-
ment. It is suggested that several sequence differences between the MT
and LXX relate to late additions of sections whose position was not yet
fixed when the archetypes of these texts were composed.
1. Josh 8:30-35
Josh 8:1-29 presents Joshua’s battle at Ai. Next comes a short section
(8:30-35) condensing into a few lines the of the building of an altar on
Mt. Ebal in accordance with Deuteronomy 27. The next chapter, chapter
9, starts with a general description:
412 CHAPTER TWENTY-SEVEN
When all the kings who were beyond the Jordan in the hill country and
in the lowland all along the coast of the Great Sea toward Lebanon,
the Hittites, the Amorites, the Canaanites, the Perizzites, the Hivites,
and the Jebusites, heard of this, 7they gathered together with one
accord to fight Joshua and Israel (9:1-2).
This description is not connected with any particular section , neither in
the preceding nor the following chapter. According to these verses, the
kings of Palestine gathered in order to fight Joshua, but no details are
specified. Possibly this remark contrasts the actions of the Gibeonites
to be described in chapter 9 with those of the other kings of Palestine
mentioned in 9:1-2. In that case the first verses of chapter 9 are indeed
connected with the continuation of chapter 9 (the Gibeonites), as in MT.
However, there is no follow-up on the remark that the kings gathered
in order to fight Joshua and Israel. It is therefore also possible that the
general statement ‘When all the kings ... heard of this’ is meant as a
conclusion to the story of Ai in chapter 8.
The opening verses of chapter 9 refer to the content of chapter 8, that
is, the story of Ai in vv. 1-29, beyond the story in vv. 30-35 about the
erecting of the altar on Mt. Ebal. Thus, originally vv. 30-35 did not
belong to the context, and that section must have been added at a later
time. One further notes that this section is unmistakably phrased in
deuteronomistic language,! and as it is based on Deuteronomy 27, it
probably originated with the deuteronomistic redactor of Joshua who
harmonistically portrayed Joshua’s accomplishing of the deuterono-
mistic law.4
Another argument for the lateness of the section is the lack of
relevance to the context, which is enhanced by the opening word In:
‘Then Joshua built an altar ...’ (8:30). This word creates an artificial
connection with the context, just like the phrases ‘in his days’ or ‘at
that time’ recognized universally as reflecting later editorial
additions in the biblical text. This section must have been added at a
place which would more or less fit the geographical and chronological
framework of building the altar on Mt. Ebal, artificially connecting two
passages. The geographical link is only roughly appropriate; Ai (8:1—
29), Ebal (8:30-35), and Gibeon (chapter 9) are all situated north of
1 See M. Noth, Das Buch Josua (HAT; 2d. ed.; Tubingen 1953) 9, 51-53.
Deuteronomy 27 itself is a composite of different commands (the erection of an altar
on Mt. Ebal in vv. 1-8 [two versions], a ceremony of blessings and curses in vv. 11-13 and a
ceremony of curses in vv. 14-26) which has been combined harmonistically in Joshua 8.
For a similar use of 18, see Deut 4:41-43, and for ‘and it occurred in that time ...’, see
Gen 38:1.
SEQUENCE DIFFERENCES 413
Jerusalem. Ai and Gibeon are close to each other, while Ebal is situated
further to the North making its mention unusual. All this leads us to
believe that 8:30-35 contains a relatively late addition to the text.
In the LXX the section about the building of the altar is found not
before chapter 9, as in MT, but after those two verses:
ie The conquest of Ai (8:1—29).
a A summarizing notice: ‘When all the kings ... heard’ (9:12)
3: The building of the altar (8:30-35).
4. The cunning of the Gibeonites (9:3 ff.).
Josh 8:30-35 has no connection with the context, and it also contains
deuteronomistic phraseology. Both of these factors lead us to believe
that this section is secondary. Furthermore, the different location of
this section in the LXX implies that its placement had not yet been
fixed. When this section was inserted into the text, it was added at the
end of chapter 8 in the forerunner of MT, and after 9:2 in the forerunner
of the LXX.
2.1 Kgs 8:12-13
1 Kgs 8:12-13 contains Solomon’s benediction spoken at the time of the
dedication of the temple: ‘Then Solomon said, ‘The Lord has set the sun
in the heavens, but has said that he would dwell in thick darkness. I
have built thee an exalted house, a place for thee to dwell in forever.”
The LXX and MT differ in several important details, and what counts
more, after these verses the LXX adds: ‘Is this not written in the book of
the Song?’ (ovk ldo atm yéypatTat Ev BLBALW THs WSAs)—‘in the book of
the Song’ (Ev BLBALw THs WSs), or Twn 1902 is probably a corruption of
the qw-n 150 mentioned in Josh 10:13 and 2 Sam 1:18.
Solomon’s benediction of vv. 12-13 occurs in the LXX toward the end
of the chapter, after v. 53.
The sequence of events according to MT is thus as follows:
12-13 Dedication of the temple by Solomon (in poetry).
14-2] Dedication of the temple by Solomon (in prose).
22=53 Solomon’s prayer.
54 ff. Blessing by Solomon.
The juxtaposition in MT of two dedications, a prayer and a blessing, is
contextually problematical. Even more difficult, the prose dedication
in vv. 14-21 starts off as if the previous poetry dedication had not yet
occurred in the immediately preceding verses. It is therefore note-
worthy that in the LXX the poetry dedication occurs after v. 53. It
seems that the combined evidence of the MT and LXX indicates that at
one point the sequence of the dedications, the prayer, and the blessing
had not yet been fixed. Furthermore, the juxtaposition of the two
414 CHAPTER TWENTY-SEVEN
benedictions is artificial and the prose dedication (vv. 14-21) as well
as Solomon’s prayer (vv. 22-53) are full of deuteronomistic elements,
and hence late.’ It stands to reason that this prose dedication as well as
the prayer are late additions inserted in the MT and LXX in different
places, as was suggested already in 1900.
These two examples suggest that in each case the difference has been
caused by the insertion of new material in two different places in the
textual witnesses. Probably the uncertainty about the placement of the
added material caused this different sequence, but it is not rmpossible
that there is more involved. It is no coincidence that in all cases the
LXX represents a more original sequence than MT.
3. Num 10:34-36
In MT wv. 34-36 read as following:
34 And the cloud of the Lord was over them by day, whenever they set
out from the camp.
35 And whenever the ark set out, Moses said: ‘Arise, O Lord, and let
thy enemies be scattered and let them that hate thee flee.’
36 And when it rested, he said, ‘Return, O Lord, to the ten thousand
thousands of Israel’ (RSV).
Vv. 35-36 contain the song which was sung when the ark traveled in
the wilderness. V. 34 appearing in MT just before this passage, occurs in
the LXX after vv. 35-36 (the sequence of the LXX thus is 35, 36, 34).
From the point of view of its content, v. 34 could indeed be placed either
before or after the song, although one would probably prefer its location
in the LXX, since the sequence 33, 35 is more natural. For v. 35 starts off
with the words ‘and whenever the ark set out ...’ and the ark, ]187, 1s
not mentioned in the previous sentence, while in Hebrew composition
the definite article should refer to something mentioned in the
immediate vicinity. Thus, the ark in v. 35 points back to v. 33 where
‘the ark of the covenant of the Lord’ is mentioned, while v. 34 of MT
disturbs the logical sequence 33, 35.
The main issue is not to determine which sequence is preferable, but
to understand the background of the different sequences. There is no
reason to ascribe the different sequence to the translator of the
otherwise rather faithful translation of Numbers. One notes that the
4 Thus C.F. Burney, The Book of Judges and Notes on the Hebrew Text of the Books of Kings
(Oxford 1903, 1918; repr. New York 1970) 112 ff.; M. Noth, Kénige (BK; Neukirchen/Vluyn
1968) 173-174.
oR. Kittel, Die Bucher der Kénige (HKAT; Gottingen 1900) 73.
SEQUENCE DIFFERENCES 415
different sequence actually involves a passage which is secondary
within the context, viz., the Song of the ark in vv. 35-36. These verses,
which stand out from their context because of their poetical nature, once
constituted a small independent unit preserving an ancient song sung at
the time of the leaving and returning of the ark. This unit was added to
its present context during one of the compositional stages, and the
artificial connection can still be recognized. The uncertainty with
regard to the place of vv. 35-36 in the LXX and MT points to the stage
when the exact location of the addition was not yet secure.
Ancient traditions further support the secondary nature of vv. 35-36.
In MT they are denoted with ‘inverted nuns’ before v. 35 and after v. 36;
these signs actually represent the sigma and antisigma of the Alexan-
drian scribal tradition, also used in the Qumran scrolls to indicate
spurious elements.® Similar evidence is included in the Talmud and
Midrash.’
4. Jer 23:7-8
Therefore, behold, the days are coming, says the Lord, when men shall
no longer say: ‘As the Lord lives who brought up the people of Israel
out of the land of Egypt,’ but ‘As the Lord lives who brought up and
led the descendants of the house of Israel out of the north country and
out of all the countries where he had driven them.’ Then they shal}
dwell in their own land.
This section (23:7-8 in MT) occurs in the LXX at the end of the chapter,
after v. 40. Like in the previous examples, the section under review
occurs just before or after a relatively large independent unit, viz., 23:9-
40. These verses comprise a collection of oracles preceded by a heading,
o°x 112, ‘concerning the prophets’ (23:9). This collection was probably
added into the existing framework of Jeremianic oracles in its entirety
either before or after the unit now known as 23:7-8.8
© See E, Tov, “Scribal Markings in the Texts from the Judean Desert,” in D.W. Parry and
S.D. Ricks (eds.), Current Research and Technological Developments on the Dead Sea Scrolls—
Conference on the Texts from the Judean Desert, Jerusalem, 30 April 1995 (STDJ XX; Leiden/New
York/Koln 1996) 41-77.
cs According to Sifre 84 on Num 10:35, these verses were denoted with supralinear and
infralinear dots. Indeed, according to Rabbi Yehuda ha-Nasi, these two verses contain a
separate unit (1ox¥¥2 150), and according to Rabbi Simeon they do not belong in their
present place. For similar statements, see b. Shabbat 115b-116a. Masekhet Soferim 6:1
continues this idea and actually indicates the place where the two verses ought to be
included, viz., in the section of the ‘traveling of the military units.’
8 The verses themselves are more or less identical to another passage, viz., 16:14—15. Also
elsewhere in Jeremiah (see Driver, Introduction, 272-273) groups of verses occur more than
416 CHAPTER TWENTY-SEVEN
5. 1 Sam 2:1-10
The Song of Hannah in 1 Sam 2:1-10 is taken by most scholars as a late
addition to the text. Indeed, the verse immediately preceding the
hymn, 1:28, is continued in the first verse after the hymn, 2:11. Further-
more, the content of the song is only loosely connected with the context.
It stands to reason that an existing psalm has been inserted in the
context just like the psalm in Jonah (Jon 2:3-10). The psalm itself
contains a contextually appropriate hymn, a song of thanksgiving by a
woman bearing a child after a long period of infertility. The key
phrase is the second part of v. 5, ‘the barren has borne seven, but she
who has many children is forlorn’ (RSV).
When this song was inserted in the original text, there was unclarity
with regard to its exact place. This can still be seen from a few words
which the LXX has in excess of MT just after the song: kal kaTé\iTlev
auTov éKet €vatrtov Kuplou (and she left him there before the Lord).
These words are more or less identical to the text of 4QSam4 just before
the song, viz., ov [ayn AWD Jinnwm (1:28), and which otherwise are
quite similar to MT 71779 ow innw (1:28). The Song of Hannah was thus
added in two slightly different places in the Vorlage of the LXX and
4Q0Sam2? (similar to MT). The MT, the LXX, and 4QSam4, also differ in
other recensional details (see Tov, “Song of Hannah”*), a fact which
strengthens the independent history of this song.
6. 1 Kings 20-21
The reversal of chapters 20 and 21 of 1 Kings in the LXX of 3 Reigns is
also related to the addition of a section to the text. The contents of
chapters 17-22 are as following in MT:
1719 Elijah saga.
20 Ahab’s Syrian war.
Zl Elijah saga continued: Naboth’s vineyard.
22 Ahab’s Scan war continued.
A mere glance at this summary reveals the difficulties inherent in the
sequence of MT. Chapter 19 is separated from chapter 21, and chapter
20 from chapter 22. The order of the LXX (21, 20) is more logical as it
has the two parts of Ahab’s war as well as the Elijah story in one
consecutive order. On the other hand, the story of Ahab’s death in
chapter 22 of MT may be considered a fulfilment of Elijah’s threats in
chapter 20, in which case MT is preferable. Preferring neither sequence,
once, indicating the complicated growth of the book, especially as some of these repeated
passages do not occur in the LXX.
SEQUENCE DIFFERENCES 417
it may be suggested that chapter 21 probably was a late intrusion.
Indeed, modern commentators have expressed their doubts about that
chapter which differs in several ways from the surrounding ones.
Among other things it portrays Elijah differently from chapters 17-19,
and Ahab differently from chapters 20 and 22.?
7. Jeremiah 10
In Jeremiah 10, the text of the LXX is shorter than that of MT and it
also has a different sequence which may be recorded as 1+4, 5a, 9, 5b, 11.
The same sequence is reconstructed for the fragmentary 4QJer® in DJD
XV. The fragment is small, but undoubtedly vv. 6-8 and 10 which are
lacking in the LXX were also absent in the scroll. It is likely that vv.
5a, 9, 5b appeared in the scroll in that sequence, as in the LXX; the
content of the scroll cannot be reconstructed in any other way.
The overall value of the LXX version of chapter 10 should be
discussed within the framework of the evaluation of the book as a
whole, but even if we concentrate on this chapter only, a few details are
clear. The translation of this and the surrounding chapters is faithful
enough in order to surmise that vv. 6-8 and 10 which are lacking in the
LXX were not omitted by the translator, but were absent in his Vorlage.
The content of these verses supports that conclusion. Chapter 10
presents a mockery on idols, reproaching their artificiality. Among
other things the chapter stresses that the Israelites are not to behave
like pagans in their attitude towards idols. The chapter also contains a
few verses which extol the greatness of the Lord God of Israel, and
precisely these verses, 6-8, and 10 are lacking in the LXX and 4QJer®. As
a result, we cannot escape the conclusion that these doxologies were
added at a late stage in the development of the chapter, stressing the
difference between the idols and the God of Israel.
This addition in MT to the original shorter text goes together with a
change in the sequence of the verses. For even though the exact course of
events can no longer be reconstructed, when vv. 6-8 and 10 were added in
the edition of MT, they caused a change in the sequence of the verses.
8. Jeremiah 46-51
The differences between the LXX and MT regarding the location of the
oracles against the nations in Jeremiah are greater in magnitude than
9 See A. Rofé, “The Vineyard of Naboth—The Origin and Message of the Story”, VT 38
(1988) 89-104. According to P.-M. Bogaert, “Le repentir d’Achab d’aprés la Bible Hébraique
(1 R 21) et d’aprés la Septante (3 Régnes 20)”, in: Elie le prophéte—Bible, tradition,
iconographie (Leuven 1988) 39-57, the Greek sequence was based on a Hebrew text.
418 CHAPTER TWENTY-SEVEN
the preceding examples. In MT these prophecies occur at the end of the
book, out of context, while in the LXX they occur aptly after 25:13.10
The LXX of Jeremiah which differs redactionally from MT also in other
matters (cf. its short text as well as sequence differences in chapters 23
[mentioned above] and 31; see Tov, “Jeremiah”*) thus reflects an
editorial difference in this matter as well. The very fact of the
different location of the oracles against the nations in the MT and LXX
reflects their secondary nature, although it remains difficult to decide
which text reflects the original position. The location of the
prophecies against the nations in the LXX is often taken as original, but
A. Rofé and G, Fischer adduced strong arguments in favor of the
secondary character of that location.!!
9. Ezek 7:3-9
Verses 3-4 and 8-9 of chapter 7 are almost identical, and there seems to
be no literary reason for their repetition. The sequence of ideas in MT is
difficult, and that of the LXX is equally problematic. For the LXX has
the two identical sections next to each other; in fact, vv. 8-9 precede vv.
3-4 in the LXX. The two identical sections probably constitute a
doublet, and one of the two parts was added in MT in one place and in
the Vorlage of the LXX in another. Although this phenomenon presents
a textual problem, the confusion must have arisen at the stage of the
literary growth of the book. See in detail Tov, “Ezekiel.”*
The analysis has attempted to establish a phenomenon in the
literary development of the biblical books. Several sequence
differences between the MT and the Vorlage of the LXX relate to late
additions of sections whose position had still not been fixed when the
archetypes of these two texts were composed. !
10-7 will bring upon that land (scil. Babylon) all the words which I have uttered against
it, everything written in this book (717 1903), which Jeremiah prophesied against all the
nations.’ Since no such oracle against Babylon is found in the vicinity of chapter 25, and
since the prophet probably had chapters 50-51 in mind, the words ‘in this book’ must have
referred to the whole collection now found in chapters 46-51. In that case v. 13 must be
understood as in the LXX, viz., with a full stop after the words 717 7902 and with the
understanding that the next words are the heading of the collection of oracles against the
nations: ‘that which Jeremiah prophesied against all the nations’ (93 29 17797" NI} WRK
oui).
a Rofé, “The Arrangement of the Book of Jeremiah,” ZAW 101 (1989) 390-398; G.
Fischer, “Jer 25 und die Fremdvolkerspriiche—Unterschiede zwischen hebraischem und
griechischem Text,” Bib 72 (1991) 474-499.
Similar phenomena have been recognized in the text of the Talmud. .See S. Friedman,
“A Critical Study of Yevamot X with a Methodological Introduction,” nya -y1n> yOXNN—
nin apna (New York 1978) esp. 305-306.
CHAPTER TWENTY-EIGHT
RECENSIONAL DIFFERENCES BETWEEN THE MASORETIC TEXT
AND THE SEPTUAGINT OF PROVERBS
At times the LXX reflects recensional stages in the development of the
biblical books differing from those reflected in MT. As a rule, the LXX
reflects an earlier stage than MT as, for example, in the case of
Jeremiah, Joshua, Ezekiel, and 1 Samuel 16-18.! Only Jeremiah is
supported by Hebrew evidence from Qumran,’ while for the others the
LXX remains the sole witness. In this context two other discrepancies
are not mentioned: the large omissions in the LXX of Job should
probably be ascribed to the Greek translator, and hence are not relevant
to the textual criticism of the Hebrew Bible? and the status of the
major differences in the Greek text of Exodus 35-40 (transpositions,
omissions) is as yet in question.
The LXX of Proverbs has not been mentioned in this context, not only
because the text cannot be assessed easily, but also because scholars
tended to ascribe its deviations from MT to inner-translational factors
rather than to its Hebrew Vorlage. When these deviations are ascribed
to the translator, they are irrelevant to the textual criticism of the
Hebrew Bible, and their main importance lies in the realm of exegesis.
However, several of these LXX deviations derived from a different
Hebrew Vorlage which often differed recensionally from MT. Such
evidence 1s presented here, but the exegetical deviations introduced by
the transJator are discussed first.
1 For a detailed discussion and bibliography, see TCU, 237-260 and TCHB, chapter 7.
2 aQjer>-4 published in DJD XV (1997).
For an analysis and bibliographical references, see C. Cox, “Elihu’s Second Speech
according to the Septuagint,” in: W.E. Aufrecht (ed.), Studies in the Book of Job (Studies in
Religion 16; Waterloo 1985) 36-53.
420 CHAPTER TWENTY-EIGHT
1. Translational factors
There is considerable evidence in the Greek translation that points to
inner-translational factors rather than a different Hebrew text behind
the differences between the LXX and MT.
1. The translation contains much evidence of contextual exegesis, in
both minor and major details.4
2. A major divergence between the two texts is the occurrence of scores
of doublets, almost all of which seem to be translational doublets of the
same verse rather than Greek translations of Hebrew doublets.? The
great number of these doublets in the Greek Proverbs is exceptional
within the Greek Bible. These doublets pertain to single words and
pairs of words, but more frequently to whole verses. As a rule, the two
elements of the doublet are juxtaposed in the same verse (e.g., 1:14; 2:21;
9:6; 15:6), but sometimes they occur in adjacent verses (1:18-19; 14:35-
15:1). Usually one of the two members of the pair of doublets is more
faithful to the Hebrew text, and the other one is free or even
paraphrastic. According to a rule laid down by de Lagarde, Proverbien,
20, the free rendering reflects the original translation, and the more
literal one a revisional rendering. While it is not impossible that the
two renderings derived both from the original translator, it is more
likely that one of them, the literal one, was added at one of the stages
of the textual transmission by a reviser who considered the original
translation too free, e.g.:
4:10 oon Hayy? 17
Kal TANOuvojoeTat ETH Cut|c cou
tva oot yévovTal ToAAal d8ol Blou
The individual elements of the Hebrew are rendered twice:
vale Kal TAN@uvOrcE TAL iva yévwvtat toda
7? ou oot
nw ETN odol
ohaeg CwHc Blou
The first set of translations is more literal than the second one.
4 The evidence is extensive. For a partial discussion, see A.J. Baumgartner, Etude critique
sur l'état du texte du ltvre des Proverbes d’aprés les principales traductions anciennes (Leipzig
1890); G. Mezzacasa, I! libro det Proverbi di Salomone—Studio critico sulle aggiunte greco-
alessandrine (Roma 1913); G. Gerleman, Studies in the Septuagint (LUA NE 52,3; Lund 1956); J.
Cook, The Septuagint of Proverbs, Jewish and/or Hellenistic Colouring of LXX Proverbs (VTSup
69; Leiden 1997).
9 See Z. Talshir, “Double Translations in the Septuagint,” in: Cox, VI Congress, 21-63.
RECENSIONAL DIFFERENCES IN PROVERBS 42]
3. Translational exegesis is visible in the addition of stichs or whole
Verses, €.£.,
a. 6:11 JO WRD TOM FWRI 727M NAN
And your poverty will come like a vagabond, and
your want like an armed man.
elt’ €utapaylvetal oor Wotrep KaKdG SdoLTIGpOG F Trevla
Kal H évéera dotep ayaboc Spope ve
Then poverty comes upon you as an evil traveller and
want like a good runner.
Although the translation is quite free, most of the elements of the Hebrew can be
recognized in the Greek. Of particular interest is the opposition created by the
translator between the kakdc OdoiTAdp0¢ and the dya0dc Spopevc, an opposition
which is further developed in a translational plus (‘11a‘ in the edition of Rahlfs):°
‘6:lla’ éav 8 doxvoc fe fEet Wotep THY 6 AuNnTéc cou, fh SE
EvSera Wotrep KaKOG Spopevc ATAUTOLOANOEL
... but if you are diligent, your harvest will come as a
fountain, and poverty will flee away as an evil
runner.
This plus at the end of the simile of the ant (vv. 6-11) further develops the theme of
v. 11 from which two elements are repeated: kakdg Spopeve and évéetra. The
previous verses mention the idle man (éxvnpdc [vv. 6,9]), and the present one, ‘11a,’
continues their idea by referring to the rewards of the opposite character, the
diokvog , a word which does not occur elsewhere in the LXX. The use in v. ‘11a’ of
words occurring in the Greek context makes it likely that the addition has been
made in Greek rather than Hebrew, and therefore the Hebrew reconstruction of
this plus by Lagarde, Proverbien, 23, Yn1 wera YR POND) Wwe a WD NII, is
unwarranted.’
b. A similar impression of composition in Greek is created by the
added simile referring to the péAtooa (bee) earlier in the chapter
(‘6:8a-c’ [not found in MT]), where the dxvypdc is told to go to the bee
and to learn from its ways. This simile is thus very close to that of the
ant found in vv. 6-11. The secondary character of this exegetical
© In the system of Rahlfs, most added stichs are denoted with a supernumerary notation
such as ‘lla’, ‘11b’, etc. Some added stichs, however, such as in 16:11 discussed below, are
not denoted in a special way.
7 The Hebrew text of 6:10-11 recurs in 24:33-34 with minor differences, and the
translation of these verses is different although dotep dya6dq Spoweve recurs in 24:34. The
translation in chapter 24 is not followed by an addition like ‘6:11a’, but on the other hand
24:34 is preceded by an added éav 6€ TovrTo Tole. Thus both the additions in “6:11a° and
24:34 as well as the one in ‘8:21a’ start with é€dv.
422 CHAPTER TWENTY-EIGHT
expansion 1s suggested by its unusual formulation as ‘or go to the bee’
which is awkward in the text after the simile of the ant.
c.11:16 9 Ww iden Ore WI 7123 Jonn Nn nwKX
A gracious woman grasps honor, and violent men grasp
wealth.
a yuvr evxdptotoc éyelpet avdpl ddEav
A gracious wife brings glory to her husband,
B Opdvoc S atulac yuvt\ utcotioa Slkata
but a woman hating righteousness is a throne of dishonor;
y tAoUTOU OKkVNpol EvSeetc ylvovTat
the idle men come fo lack wealth,
5 ol 56€ dvépeton EpelSovtTat TAOUTW
but the diligent support themselves with wealth.
In the MT of 11:16 Jn nwx (yuvt) Eebxdptaotoc) is contrasted with o°x-ny; both of
them ‘grasp’ something different. However, the translator probably did not
understand the exact meaning of the first stich—hence his unusual rendering—nor
did he realize the exact nature of the parallelism—hence his addition of two
stichs. These added stichs (8, y) provided oppositions to the two stichs
representing MT (a,8). To the yuvy evydptotog (a) the LXX added stich 8
concerning the yuvr) picotica. The content of this added stich has close connections
with the wording of the translation elsewhere, so that it was probably added by
the original translator himself. For the first phrase in stich 8, cf. 8p6vog alabtyoews
(a throne of knowledge) in 12:23. The ‘throne of dishonor’ is probably meant as
the opposite of the ‘throne of honor’ (7123 xoOD), mentioned in 1 Sam 2:8; Isa 22:23;
Jer 14:21; 17:12. It reminds one also of the @pdvocg dpyfc¢ (Prov 16:12) used in
connection with dtKatoovv7 (as here), as well as of similar phrases (20:28; 25:5;
29:14). For the last phrase of that stich cf. 13:5 M6yov G&tkov proet Slkatocg.
To stich 6 reflecting MT the translator added stich y as contrast. This stich
creates an opposition between TAovTou, not obtained by idle men, in y, and TAoUTw,
obtained by the diligent, in 8. At the same time, the wording of this plus is based on
the vocabulary of the ‘canonical’ section, 6:6, 11, as well as of the added ‘6:11a’
dokvoc ... évSera 8
d.12:11 92) 70m op ATM aN? yaw? inNTK Tay
He who tills his land will have plenty of bread, but he
whose pursuits are empty has no sense.
11 a 6 EpyaCopevoc Thy EavTod yhv EutAnoOhoeTat dpTwv
He who tills his land will be satisfied with bread,
B ol &E Sidbkovtec pdtata évdectc hpevisv
8 Asa result, the attempt of some scholars to reconstruct a Hebrew Vorlage of this Greek
plus seems unwarranted. Note, e.g., BH: non a? xy 77 Aw? NRW AWN 777)? NOD.
RECENSIONAL DIFFERENCES IN PROVERBS 423
but they that pursue vanities are void of understanding;
‘lia’ y = 6g €otiv Hdve Ev olvwv BatpiBatc;
he who enjoys himself in amusements of wine
5 év Totc EauTtot dxyupwyao iy kaTtadeler attulav
will leave dishonor in his own strongholds.
The Hebrew verse presents an opposition between 1n97N 71y and 07/2" 770; v.
‘lla’ of the LXX adds a parallel to the latter.
dxupuuaoiv of the added stich 6 is based on dxUpwpa occurring in v. 12. From
the fact that it occurs in the next verse, rather than a preceding one, one might
conclude that the Greek addition was made on the basis of an already existing
translation.
Stichs y& continue the train of thought of stich B, even though the verse is
phrased in the singular. They probably elaborate on the theme of natata
mentioned in stich 8. The addition uses dtitav from the context (v. 9) and this
word also features in the plus in 11:16 (see above). Elsewhere, too, dtula is a
favored word of the LXX of Proverbs. For the reference to the drinking of wine cf.
also Prov 23:20; 31:4.
e.17:21 9 232 738 Mawr 87177 AMIN? 7°0D 17°
He who fathers a stupid son makes sorrow for himself
and the father of a fool has no joy.
a KapSla S€ ddpovog ddtvn TH KEKTYLEVOD AUT
The heart of a fool is grief to its possessor.
B otk EvdpatveTat TaTrp ET Uld dTAaLdsevTH
A father rejoices not over an uninstructed son,
y uldg S€ Ppdvipoc Evdpalvelt pNnTEpa avTov
but a wise son makes his mother happy.
The meaning of the Hebrew verse is lost in Greek, probably because the translator
read 15 instead of 19°. For the phrase, cf. 12:23 o°7°09 17 - kapSla 8 ddpdvav, and
15:7 (for a similar change, see the LXX of 17:10). Possibly because of the lack of a
good parallelism between stichs a and B, stich y was added as an antithetical
parallel to the second stich. At the same time, stich y was added because of the
association with the Hebrew and Greek text of 10:1 (cf. also 15:20; 23:24), where
the same rare word 711m is used as here.
The list of these inner-translational pluses is long. For similar pluses
of the LXX, see ‘4:27a’ (note the expansion on the theme of ‘right’ and
eft’ found in the MT and LXX of v. 27; v. ‘27b" contains a double
translation of v. 26); ‘7:1a’; ‘8:21a’; ‘9:12a-c’, ‘9:18a-d’; ‘10:4a’; ’12:13a’;
*13:13a’; ‘17:16a’; ‘18:22a’; 19:7; ‘22:14a’; ‘24:22a-e’; ‘25:10a’; °27:20a’;
424 CHAPTER TWENTY-EIGHT
‘27:21a’; ‘28:17a’. This list also includes cases of additions made on the
basis of verses from other books:?
£1309 JT oywr an naw? OP" 7X TX
The light of the righteous shines brightly, but the lamp
of the wicked will be put out.
duc Sikalorg 6a tavtde
duc 6€ doeBuv oBévyvuTat
The righteous always have light, but the light of the
ungodly is quenched.
To the opposition between &ikalorg and doeBav in this verse, v. ‘9a’ adds a
similar opposition:
‘13:9a’ wuyal SdALa TAaviivTaL é€v Guapttatc Slkatot 5é ol kT{pou-
ov Kal €dEWOoLV
Deceitful souls wander in sins, but the righteous have
pity and are merciful.
The second part of this addition may be based on Ps 37(36):21 (cf. also 111
[112]:4):19
Ps 37:21 7M.) PoTN) OFWs 8?) YHA ANY
The wicked borrows and does not pay back, but the
righteous is generous and gives.
Ps 36:21 SavelCetat 6 duaptwrdc¢ kal ovK dtotetoet 6 5€ S{kalocg
olkt{per kal &.8ot
The sinner borrows and does not pay back, but the
righteous has pity and gives.
For a similar addition in the context, see Prov 13:11 8Lkatog olktlpet kal KLypd
(the righteous has pity and lends).
2. Text-critical evaluation
1. The preceding section provided ample evidence of changes made
either by the translator or during the course of the textual transmission
of the translation. From the outset it thus would seem reasonable to
ascribe all major differences between the translation and MT to these
factors. However, there are indications that beyond the afore-
mentioned instances there are also major differences between the two
9 In addition to the below mentioned examples, see 1:7 (cf. Ps 111:10); 3:16 (cf. Isa 45:23
and Prov 31:26 [see below]); ‘26:11a’ (cf. Sir 4:21).
At the same time, the origin of the idea of the wandering souls (of the living or the
dead?) as in Proverbs is not clear, although one is reminded of Wisdom 17:1 aral8eutot
wuxal étAavHAOnoav. and Prov 21:16: dvyo mAaviipevog €& 5800 SiKaLoobunc.
RECENSIONAL DIFFERENCES IN PROVERBS 425
texts deriving from a different Hebrew Vorlage used by the translator.
This situation makes the text-critical evaluation of the LXX of
Proverbs very difficult.
a. The translation of 3:16, referring to Wisdom, contains several
details beyond MT. After on" 77x (long life) it adds kat ET Curie (= mw)
ovn [cf. v. 2]) and after TAottog kat 56€a, it adds two stichs (’3:16a’):
‘3:16a’ a €k TOU OTépaTOSE aUTH< ExTopEevETAat SiKaLOOUVN
Out of her mouth proceeds righteousness,
B vdpov Sé Kal EdAE€ov ETL yAWaons dhopet
and she carries law and mercy upon her tongue.
Stich a is based on Isa 45:23 P78 Dn NB" (from my mouth righteous-ness goes
forth), where the LXX uses a different verb, €€eAevoeTar. Stich B provides a more
literal version of Prov 31:26 anw? 2y Ton nvm (and the teaching of kindness is on
her tongue) than the LXX ad loc. :
31:25 .. EvVVOUWG. Kal TdELY €aTtelAATO TH yAWoon auTic
(... and lawfully?). And she commanded order to her
tongue.
Although inner-Greek activity cannot be excluded, the inner-translational
differences between the translations in ‘3:16a’ on the one hand and 31:26 and Isa
45:23 on the other make it likely that the plus in Prov ‘3:16a’ did not derive from
inner-Greek activity. Rather this plus is based on an expanded Hebrew text (779n
aw) 9y Fon NNN APTN XxXN ?),
b. The plus in ’3:22a’ is more or less identical with the text of 3:8:
LM
‘3:22a’ €otat dé taovc taic oap&l cou
Kal eTLUédELa Tote cote daTEOLG
It will be healing to your flesh and safety to your bones.
3:8 MEY? “PW FAW? “IN MND
It will be healing to your flesh and refreshment to your
bones.
TOTE LaoLe EoTat Tus GWYaTl Gou
Kal émiuéde€La Tote daTéolG dou
In both cases the Greek text occurs after negative commands (7 ‘Nn 2x - ph) ToOL; 21
1? 2X - ph Tapappurc). It is not likely that the text of ‘3:22a’ has been repeated on
the inner-Greek level, since the two translations differ. Rather, the discrepancies
between the two texts most likely derived from different translations of the same
Hebrew text (note the differences between Téte [8] and Sé [‘22a’], the different
426 CHAPTER TWENTY-EIGHT
rendering of 77wW?, TH odpatl cov [8], Taic capEl cov [‘22a’],!! and the differences
between oov [8] and cot¢ [‘22a’].) In that different Hebrew text the verse may have
occurred twice, and in both places it suited the context.
c. The same reasoning obtains regarding the repetition of the
following verse:
2771 or 75> an yin x? 7D (ann ara B9ANN 2x)
For you do not know what a day may bring forth,
ov yap yLvidoKete TL TEEE TAL T) ETLOUOA
for you do not know what the next day will bring forth.
3:28 ...0U yap olSac Tl TéEeTaL Ff ETTLOGCAa
The contexts in which the verse occurs in both places are similar, in both cases
after Inn in the preceding stich. In 27:1 the Greek has an equivalent in MT, but it
has none in 3:28. The occurrence of this verse in 3:28 probably does not represent
an inner-Greek repetition (note the differences between the verbs in the two Greek
versions). Rather, it reflects a Hebrew text in which the verse occurred twice.
Since MT itself contains several instances of recurring verses (see n. 7), it is not
surprising that the Vorlage of the Greek contains additional instances of recurring
verses.
2. Major differences between the two texts are visible in the trans-
positions of verses and groups of verses. Rahlfs denoted these verses as
supernumerary pluses (‘12a’, etc.), as in the preceding examples, but
actually they represent transpositions, often coupled with pluses and
minuses. The numbering in the edition of Rahlfs thus creates a
misleading tool for its investigation.
a. The main example of this phenomenon is found in the verses at the
end of chapter 15 and the beginning of chapter 16. The sequence of the
verses in the LXX is as follows according to the numbers of MT:
15:1-27
16:6 (Rahlfs: ‘15:27a’)
bo-28
1627 (Rahlfs: ‘15:28a’)
i529
16:8 (Rahlfs: ‘15:29a’)
16:9 (Rahlfs: ‘15:29b’)
15:30
1] Both Greek words are known as translation equivalents of the same words, even if
the exact equivalent in this verse is not clear (qwa or 1W = 1xw). Elsewhere in the LXX of
Proverbs, owua reflects axw (5:11; 11:17; instead of o-qwa in 25:20 the LXX read another text,
either (wa or WNW) as well as Ww (4:22; 5:11; 26:10 [?]). Elsewhere, odp— reflects w2 (passim)
and axw (Mich 3:2, 3).
RECENSIONAL DIFFERENCES IN PROVERBS A? 7
15:32,.33 (note omission of v. 31)
16:2 (?)12 (note omission of v. 1)
16:5 (note omission of v. 3 and transposition of v. 4)
‘16:7’ first stich of the LXX (note omission of v. 6)
16:7 first stich, represented as the second stich of 16:7 in the LXX
16:8 (differing from v. 8 of MT)—in other words, the
greater part of vv. 7-8 of MT lacks in the LXX
16:4 (Rahlfs: 9)
16:10 ff.
The reason for these major changes is not connected with the textual
transmission, as suggested by Lagarde!¥ nor with the disorderly status
of the manuscript(s) from which the translation was made.!* Rather,
the two texts represent recenstonally different editions. The sequence of
most sayings in these chapters is loose, and as each one is more or less
independent, two different editorial traditions could have existed
concerning their sequence. One notes especially the transposition of
several verses of what is now chapter 16 to what is now the end of
chapter 15; one also notes the change of position of 16:4. These
phenomena are coupled with the omission of 15:31; 16:1,3, and the
replacement of 16:6-8 of MT with two different Greek verses (numbered
16:7-8 by Rahlfs). Further, 15:31 (‘11 nynw jx), lacking in the LXX,
could have been added secondarily in the edition of MT as an appendix
to the previous verse dealing with ory Wx» and 721 AYInw. The first
eleven verses of chapter 16 in MT display a certain principle (occurrence
of the name of God in all verses except for vv. 8 and 10), but this
situation does not necessarily render that version preferable to that of
the LXX, where such an editorial principle is not visible. Furthermore,
the type of parallelism of the verses in the arrangement of MT does not
make it a more coherent unit than that of the LXX.
b. The sequence in chapter 20 is as follows in the LXX (according to
the verse numbers of MT):
1-9
20-22 (Rahlfs: ‘9a-c’)
12 Tt is not certain that the verse which is denoted by Rahlfs as 16:2 indeed represents
16:2 of MT, as it also presents elements that could be taken as reflecting 16:4.
3 Lagarde, Proverbien, 51 suggested that the text of chapters 15 and 16 was written in
adjacent columns and that the translator wrongly read the text horizontally rather than
vertically. However, de Lagarde took into consideration only the transposition of the verses
from chapter 16 to chapter 15, and not the other phenomena in the translation (omissions,
additions), and therefore his solution is less plausible.
4 Thus Baumgartner, Etude critique, 149 (cf. n. 4 above).
428 CHAPTER TWENTY-EIGHT
10-13
23-30 (note omission of vv. 14-19)
As in the preceding case, there is no logical connection between the
verses, and both sequences are possible. Editorial rather than scribal
factors must have determined the different sequences, as this is also
coupled with an omission (vv. 14-19). Toy! ascribed these different
sequences to ‘accident or scribal caprice.’
c. The sequence in chapter 17 is as follows in the LXX (according to
the verse numbers of MT):
1-16
19b (Rahlfs: ‘16a’)
20b (Rahlfs: ‘16a’)
17-18
19a
20 including a translation of v. 20b (also translated in the LXX
of v. ‘16a’)
21-28
d. The sequence in chapter 31 is as follows in the LXX (according to
the verse numbers of MT):
1-24
26 (Rahlfs: 25)
25 (Rahlfs: 26)
27
26a (Rahlfs: ‘28a’)—a second translation is found in ‘3:16a’
28-31
e. The same explanation applies to major differences in sequence
between the various segments of the book in chapters 24-31. According
to their headings, the following eight collections of proverbial
material are recognized in the book of Proverbs according to MT:!¢
I T1918 (‘The proverbs of Solomon’)
i 10:1—22:16 (‘The proverbs of Solomon’)
III 22:17—24:22 (‘The words of the wise’)
IV 24:23-34 (‘Also words of the wise’)
V 2929 (‘These are also proverbs of Solomon which
the men of Hezekiah king of Judah copied’)
IS CH. Toy, The Book of Proverbs (ICC; Edinburgh 1899) 388.
16 Toy, Proverbs, vi subdivides MT into five consecutive sections. Our own understand-
ing is closer to that of W. Frankenberg, Die Spriiche (HAT; Gottingen 1898) 2-5 who
mentions eight subgroups and Eissfeldt, Introduction, 472, who speaks of seven sections. Of
the commentators, only Frankenberg, pp. 10-11 paid detailed attention to the sequence of
the LXX, the logic of which he tried to explain.
RECENSIONAL DIFFERENCES IN PROVERBS 429
VI 30 (‘The words of Agur’ [and other sayings])
Vil 31:1-9 (‘The words of Lemuel’)
VIII 31:10-31 (an acrostichon about the virtuous woman)
This description of the contents of MT is based on explicit headings in
that text, but at least in two cases these headings may be misleading.
Chapter 30 is represented as ‘the words of Agur’ (and other sayings)
since v. 1 contains the only heading in this chapter. However, most
commentators doubt whether all of the verses in this chapter should be
ascribed to a collection of ‘the words of Agur.’ Indeed, the nature of vv.
15-33 (numerical sayings) differs from that of the first 14 verses, and
probably the real ‘words of Agur’ comprised even less than 14 verses.
Therefore, when representing here and below ‘the words of Agur’ as one
section, this formal approach may be misleading. Likewise, not all of
chapter 31 should be ascribed to ‘the words of Lemuel,’ and its second
part, an acrostichon about the virtuous woman, should be considered a
separate unit.
The sequence of the LXX can be described as following according to
the sections and numbers of MT:
I-III 1:1—24:1-22
VI, part 1 30:1-14 (‘The words of Agur,’ first part)
IV 24:23-34 (‘Also words of the wise’)
VIij.part 2 -30:15-33 (‘The words of Agur,’ second part)
VII 316129 (‘The words of Lemuel,’ first part)
V 25-29
VIII 31:10-31 (an acrostichon about the virtuous
woman, formally representing ‘the
words of Lemuel,’ second part)
In other words, the LXX separates between the two parts of section VI
(‘The words of Agur’) and of chapter 31 (VII [‘the words of Lemuel’] and
VIII [the acrostichon of the virtuous woman]). Furthermore, it reverses
the internal order of sections IV, V, VI and VII, part 1.
When turning to a comparative analysis of the sequence in the MT
and LXX, neither one of the two systems should be preferred. The
connection between the sections is such that both can be equally correct.
From the outset the juxtaposition of sections III and IV, as in MT, is to
be preferred to the arrangement of the LXX as III contains ‘the words of
the wise’ and IV ‘also the words of the wise’ (thus Frankenberg [n. 18]
who considers IV a ‘Nachtrag’ to III). However, one could also argue
against the arrangement of MT. For why should collection IV need a
separate heading if both it and the previous collection contain “words
of the wise’? Therefore, the arrangement of the LXX has more to be
430 CHAPTER TWENTY-EIGHT
recommended than that of MT, since the separation of IV from III
requires a separate heading for IV, as in the LXX.
The separation in the LXX between the different sections of ‘the
words of Agur’ and ‘the words of Lemuel’ is contextually no better or
worse than their juxtaposition in MT. One should remember that both of
these collections are composed of at least two segments whose contents
are not necessarily connected. Thus not all of the sayings in chapter 30
should be considered as ‘the words of Agur.’ In any event, vv. 15-33
(various numerical sayings) are set apart, and could certainly be placed
elsewhere. Likewise, chapter 31 is composed of different segments; its
second part, an acrostichon about the virtuous woman, is not connected to
the first part, ‘the words of Lemuel,’ and could therefore be placed
elsewhere, as it is in the LXX. In the arrangement of the LXX the second
part of ‘the words of Agur’ (VI, part 2) has no separate heading, and
therefore belongs, as it were, to section IV (‘also words of the wise’);
contextually this arrangement is equally as good as that of MT. On the
other hand, both Agur (VI) and Lemuel (VII) are described as ‘of
Massa’ (‘the Massaite’), so that their juxtaposition in MT, at the end of
the book is preferable to their separation in the LXX. However, even in
MT the ‘words of Agur’ are not really juxtaposed to ‘the ‘words’ of
Lemuel,’ since the second part of chapter 30 actually does not contain
sayings of Agur.
In this description, the arrangement of MT has been compared with
that of the presumed Vorlage of the LXX, beyond the understanding of
the translator. For the translator often misunderstood the nature of the
headings. Thus the following headings have been misunderstood in the
translation:
24:23, U7: UDWNI OD 197 ONIN? APN
These also are words of the wise. Partiality in judgment
is not good.
tavta Sé€ dé yw vptv toic codote étLtytviioketv aldetabar
Tpdawtov €v Kploet OU KadV
And these things I say to you, the wise men, to know: it is
not good to respect a face in judgment.
The heading has been taken as an integral part of the sentence.
30:1 NWT TP* 72 Wx 7374
The words of Agur the son of Jakeh of Massa.
Tove €ove Adyouc vLE HoBHOntL kal SE€EduEevoc avtove
LLETAVOEL
my son, fear my words, and receive them and repent
RECENSIONAL DIFFERENCES IN PROVERBS A3]
The proper name Win has been taken as a verbal form, and the first word has been
read as "27.
311 (ON INNO" WR) Xwy 2p PNW? -717
(The words of Lemuel, king of Massa [which his mother
taught him].)
ol €ol Adyou ElpnvTat UTd Beot, Baotkéwe ypnuaTiopdc
(My words are spoken by God, an oracle of the king)
Like in 30:1, the first word has been read as ~727, and the proper name Lemuel has
been separated into two parts. These changes brought about further changes in the
translation.
3. Another indication of a different Vorlage is the fact that in
various instances the text of the LXX is shorter than that of MT: 8:29a,
33; 11:4, 10b,11a; 15:31; 16:1,3; 18:23-24; 9:1-2; 20:14-19; 21:5, 18b; 22:6:
23:23. The number of these examples is too large to assume a scribal
phenomenon (parablepsis).
3. Conclusion
[It seems that the translation was made from a Hebrew copy of Proverbs
which differed recensionally from that of MT. These differences
consisted of major and minor differences in sequence as well as
differences in pluses and minuses. If the interpretation of these
differences is correct, we have gained further insights into the history
of the growth of the book of Proverbs. At a relatively late time the
different editorial stages of the growth of the book were still reflected
in the texts.
When Proverbs was translated into Greek, presumably in the second
century BCE, a scroll was used that contained an editorial stage of the
book differing from that now contained in MT. Such an understanding
parallels views developed previously regarding other biblical books.
This view does not imply that the editorial changes were made as late
as the time of the Greek translation, but that at that time, in a
geographically remote center of Judaism, such early scrolls were still
available.!?
17 This view had already been suggested by Swete, Introduction, 241, although he still
allows for the possibility that the translator himself may have been involved in the
changes. Our own views are more in agreement with those of Mezzacasa (n. 4) 2-3;
Eissfeldt, Introduction, 472; and S. Ahituv, “Proverbs,” Encyclopaedia Biblica 5 (Jerusalem
1968) 554 (Hebrew). The latter three views mention the possibility of recensional
differences between the MT and LXxX, although none goes into detail.
CHAPTER TWENTY-NINE
DIFFERENT EDITIONS OF THE SONG OF HANNAH AND OF ITS
NARRATIVE FRAMEWORK
1. Introduction
The differences between MT (with which T,! S, and V more or less agree)
and the LXX? in the Song of Hannah are mentioned in the commentaries
and in several monographic studies of that poem.? The sources differ in
many small details, as well as in major ones in vv. 1, 2, 6, 9, 10. These
major discrepancies consist of differences, omissions, and additions
(when using these terms, MT is taken as point of departure without
taking a stand regarding the originality of the readings of that text). As
far as I know, the differences between MT and the ancient versions of the
Song of Hannah and its narrative framework have not been discussed in
a monographic treatment,’ with the exception of Walters, “Hannah and
Anna” (on the relation between the MT and LXX); nor have the
differences between MT and 4QSam#? been discussed. When deviating
from MT, this scroll often agrees with the LXX and/or LxxLuc (see Tov,
“Qumran,”* and “4QSam?”*). The differences between the Qumran scroll
and MT have been put forward in Cross, “New Qumran Fragment,”
! See DJ. Harrington, “The Apocalypse of Hannah: Targum Jonathan of 1 Samuel 2:1-
10,” in D.M. Golomb (ed.), “Working with No Data,” Semitic and Egyptian Studies Presented
to Thomas O. Lambdin (Winona Lake, IN 1987) 147-152.
2 The Old Latin version is more or less identical with the LXX. See in detail P.A.H. de
Boer, “Confirmatum est cor meum—Remarks on the Old Latin Text of the Song of Hannah
1 Samuel 11 1-10," OTS 13 (1963) 173-213; idem, “Once Again the Old Latin Text of
Hannah’s Song,” OTS 14 (1965) 206-213.
3p, Dhorme, “Le Cantique d’Anne (I Sam. JI, 1-10),” RB 16 (1907) 386-397; G. Bressan, “Tl
cantico di Anna (1 Sam 2,1-10),” Brblica 32 (1951) 503-21; 33 (1952) 67-89; J.T. Willis, “The
Song of Hannah and Psalm 113,” CBQ 35 (1973) 139-154; M. Phiflonenko, “Une paraphrase
du Cantique d’Anne,” RHPHR 42 (1962) 157-168; Stoebe, Samuel; P.A.H. de Boer, “Einige
Bemerkungen und Gedanken zum Lied tn 1 Samuel 2,1-10,” in: Beitrage zur alttestament-
lichen Theologie, Festschrift fiir Walther Zimmerli zum 70. Geburtstag (Gottingen 1977) 53-59;
D.N. Freedman, “Psalm 113 and the Song of Hannah,” Erlsr 14 (1978) 56*-69*; Tournay,
“Cantique.” Further monographs are mentioned in Lewis, “Hannah.”
4 These details are not discussed by S. Pisano, S.J., Additions or Omtsstons in the Books of
Samuel (OBO 57; Freiburg / Gottingen 1984).
434 CHAPTER TWENTY-NINE
Ulrich, Samuel, the notes of BHS (P.A.H. de Boer) which in Samuel] are
more extensive and more cautious than in the other books in that edition,
the textual notes on the New American Bible,? and McCarter, Samuel. What
has not been sufficiently recognized is that the three different texts of the
Song of Hannah do not merely reflect scribal differences such as are
created in the course of the transmission of any text, but reflect three
different editions (recensions) of this Song and its narrative framework.
That this is the case was, however, suggested long ago with regard to the
MT and LXX. Wellhausen, Samuel, 42 referred to the different position of
the Song of Hannah in MT and LXX, and Driver, Samuel, 22 on the MT
and LXX of 1:28 and 2:11, while Barthélemy, CT 1, 144-145 referred to
2:8-9, The difference between scribal and editorial activity is difficult to
define and even scholars who agree in principle that there is a category
of editorial differences often do not agree with regard to individual
instances. When using the terms editorial or recensional, we refer to
readings which presumably were created before the completion of the
composition. When these readings were created, the biblical composition
was still fluid, so that generations of editors allowed themselves to
change that composition. The main focus of this study is the Song of
Hannah and the surrounding verses, but in a way, the history of that
Song cannot be separated from that of the surrounding chapters. If
different editions of the Song of Hannah are assumed, evidence for such
editions should also be visible in other chapters in the book of Samuel.
2. An analysis of the major differences
The view that different editions of the Song of Hannah are reflected in
the textual witnesses 1s based on an analysis of the textual data.
a. The position of the Song of Hannah (1 Sam 1:28; 2:11)
The Song of Hannah is placed in a slightly different location in the three
textual traditions as shown by a comparison of the verses before and
after the Song in 1 Sam 1:28 and 2:11:
1 Sam 1:28
LXX 4QSam? MT
> m>/]innwm ow w[aryme me? ow wn
1 Sam 2:11a
LXX 4QSam?2 MT
> The Holy Bible, I. Samuel to Maccabees—New American Bible (1968); cf. also: Textual Notes
on the New American Bible (Patterson, NJ. [n.d.]).
An alternative reconstruction would be 1[M1m, as suggested by Wellhausen, Samuel, 42
for the LXX of 2:11.
SONG OF HANNAH 435
Wie? 152 Ow Adaivm So? Wa: PUA aa oe 2
anna 2M
Kal KaTéAutrev’ abtov
EKEL EVWTILOV KUpLou Kal aTmev els Apuaiat
The main actions described in 1:28 and 2:11, leaving Samuel at the
temple and the bowing before the Lord, are ascribed to different persons
in the various textual traditions or are not mentioned at all (the bowing
before the Lord is lacking in the LXX), as will be discussed in the next
section. These actions take place at different points in the story.
According to MT, an unidentified person bows before the Lord prior to
Hannah’s Song. In a similar way, according to 4QSam?, Hannah
prostrates herself before the Lord before the Song, and at that point she
leaves Samuel at the temple. On the other hand, according to the LXX,
Hannah leaves Samuel at the temple after the Song. Since the actions
themselves are more or less identical, the data could also be presented as
the insertion of the Song at two different positions, according to 4QSam?
after Hannah’s actions, and according to the LXX before these actions.
MT resembles the scroll inasmuch as it describes an action before
Hannah’s Song, but it differs from the LXX and 4QSamé since it ascribes
the actions to Elkanah. The insertion of the Song at two different
locations in the context may indicate the late addition of that Song in the
history of the growth of the first chapters of Samuel since the Song did
not belong to the first layer of the text. When it was inserted into the text,
it was inserted in a slightly different place in some manuscripts. A
similar explanation applies to the different place in the textual traditions
of the Song of the Ark (MT: Num 10:34-36), the pericope on the building
of the altar (MT: Josh 8:30-35), Solomon’s benediction for the dedication
of the temple (MT: 1 Kgs 8:12-13), the story of Naboth (MT: 1 Kgs 20-21),
and the oracles against the foreign nations in Jeremiah (MT: chapters 46-
51).8
7 For this verb as well as for dmm\Gev, we follow (against Rahlfs) the text of B and a few
other sources, disregarding the matin evidence of the Greek tradition, which has plural
forms (kaTéXe)itrov, aTjA6ov). The slight difference between the two readings is scribal and
cannot be ascribed to revisional tendencies relating to MT, since there is no equivalent for
these words in MT. For the evaluation of the inner-Greek differentiation between the two
textual traditions, the principle of the lectio difficilior is invoked. Since Hannah and Elkanah
were together tn Shiloh at the second visit (see below), it is more logical to ascribe this
action to both of them, so that the plural form of the majority of the Greek tradition should
be taken as an inner-Greek correction. The more difficult singular form in the LXX of 2:11 is
assumed to be original, and it is this form which agrees with the text of both MT and
AQSam® in 1:28.
8 Thus already briefly Wellhausen, Samuel, 42. See further Tov, “Sequence.”*
436 CHAPTER TWENTY-NINE
The evidence of 4QSam¢ is only partly known. In 1:28b, before the Song
of Hannah, the scroll mentions the actions which appear in the LXX after
that Song. Although the verses after the Song have not been preserved in
4QSam/?, a calculation of the available space easily enables the inclusion
of v. 1la.?
b. 1 Sam 1:24, 25, 28; 2:11: the dramatis persona(e) during the third visit to
Shiloh.10
There are major differences between the textual sources regarding the
conception of the dramatis persona(e) during the third visit to Shiloh.
The analysis of these differences is hampered by textual complications in
MT, the difficulty of reconstructing the Vorlage of the LXX, and the
fragmentary state of preservation of the Qumran scroll. Nevertheless the
main facts are clear:
According to MT, Hannah is the main person acting in 1:24-28. The
first words of v. 24 (any 177yM1) make it clear that she came up to Shiloh
with Samuel, but seemingly without Elkanah, and it is she who acts in
vv. 24-28. However, an unidentified male person is mentioned at the end
of the chapter in v. 28b a7? ow innwy, ‘and he bowed there before the
Lord.’ From the immediate context it is unclear who is referred to,
although on the basis of the earlier verses (cf. v. 21) it is likely that
? On the basis of the columns containing the text of chapters 3, 4, and 5 of 2 Samuel (see
photograph PAM 43.115), the column length of this scroll may be calculated at 43-44 lines
(in the case of the column starting with 2 Sam 3:23, the lines average 40-45 spaces). The first
almost completely preserved column on photograph 43.115 preserves the top margin as
well as remnants of 34 lines containing the text of 2 Sam 3:23 until 4:4. Since the next
column preserves likewise a top margin immediately followed by the text of 2 Sam 4:9, the
bottom of the first column must have contained the text of vv. 5-9. That text is
reconstructed as an additional 9-10 lines, bringing the total number of lines for that column
to 43-44. By the same token column I of the scroll would have contained an additional 7-8
lines after the 36 partially preserved lines of this column. The next column, col. I], starts
with 1 Sam 2:16 and continues with vv. 13b, 14, 17ff. It is therefore difficult to assess the
exact evidence of the scroll, but it seems that the unusual text at the beginning of this
column, which is in the nature of a duplication with changes, should be disregarded in the
present analysis. Thus at the end of col. Il there was ample room for v. 11a, more precisely
for 1 Sam 2:11-16, partly duplicated at the beginning of col. III. For that column also 43-44
lines should be reconstructed.
0 The first visit is described in 1:3-18, the second one in 1:21-23, and the third one in
2:24-2:11. At the end of the first and third visits, Elkanah and/or Hannah retum to their
home (1:19; 1:28; and 2:11), while such a formal statement is lacking at the end of the
second visit. If the second and third visit are regarded as one event, some of the problems
described here are resolved, since in that case Elkanah is mentioned explicitly, though at a
great distance from v. 23. This assumption is, however, difficult, since it implies that
Elkanah would have to wait a very long period in Shiloh, about which nothing is said in
the text. The argumentation below is not affected by this assumption.
SONG OF HANNAH 437
Elkanah is meant.!! That this is indeed the case becomes clear from the
first verse occurring immediately after the Song of Hannah, viz., 2:11,
‘and Elkanah went to his home to Rama.’ If according to this verse,
appearing immediately after the Song of Hannah, Elkanah returned to
his home, he must have been away from his home, in Shiloh, so that the
subject of the verb in 1:28 has to be Elkanah, even if he has not been
properly introduced, so to speak, in the account of the third visit to
Shiloh (cf. n. 17). The reason for the lack of explicit mentioning of
Elkanah in 1:28 becomes clear from an analysis of the preceding verses,
in which apparently a textual mishap had occurred.!* Whatever was the
background of the phrase in MT in 1:28, the text of MT is problematic
since its subject is not disclosed. S and V have a plural form (cf. v. 19 in
the various witnesses), but in these translations this form probably
reflects a contextual harmonization.!° However, the difficulty in v. 28b is
not created by a textual problem of conflicting verbal forms, but is part of
a discrepancy between different editions of the Song, now reflected in
the various textual witnesses.!4 What exactly happened with the text of
MT is unclear, but 1 Sam 2:11 describes Elkanah as being present in
Shiloh at the time of the third visit to Shiloh. While these words are not
represented in the LXX, they probably formed part of the original design
of the edition reflected in MT, as becomes clear from an analysis of a
section appearing before the last word of v. 24 in the LXX and 4QSam4.
That section is lacking in MT and was probably omitted erroneously
from it through a special! type of homoioteleuton (—7y17) 1vin). The
very fragmentarily transmitted text of 4QSam can be reconstructed well
in accordance with the LXX, with which it agrees:
Kal TO ratddptov wet abTadv Kal Tpoohyayov évuTLav Kuplou kal éadaEev 6
TaThp avTov Thy Obvolav ty étroler €E he pav els hpépas Ta Kuplw. kal
Tpoanyayev TO tratddpiov
mie) an oon wy] werd njara [mx rox unw mi 1D? way ooy] vim
[avin nx RIM
11 On the other hand, according to McCarter, I Samuel, 58, MT refers to Eli. Walters,
“Hannah and Anna,” 401, thinks of Samuel. Because of the unclear context, a case can be
made for both of these persons, but in view of 2:11 (see below) and of the text omitted from
MT in 1:24 (see below), only Elkanah can have been intended here.
12 Tt is not likely that Elkanah’s name has been omitted on purpose as part of a narrative
technique (Walters, “Hannah and Anna,” 400).
3 Likewise NJPST: ‘And they bowed low.’ A textual note in that translation refers to b.
Ber. 61a implying that Elkanah was there.
4 Cf. further the Ketib in Gen 43:28, where wnnw") represents a plural form.
> The two identical words are both contained in MT.
438 CHAPTER TWENTY-NINE
Apart from the suggested homoioteleuton, it is difficult to account for
the present shape of MT. There is, for example, no reason to assume an
intentional omission of this phrase in MT.!© Besides, the information in
the plus of 4QSam@ and the LXX is needed for the understanding of the
surrounding verses in MT, so it is reasonable to suggest that it was
omitted erroneously from that text. This plus mentions Elkanah, so he
would be the subject of the verbs in the singular in v. 28. It is he who
brings the boy to Eli (this action runs parallel to Hannah’s leaving the
boy in the temple in 1:28 in 4QSamé@ and in 2:11 in the LXX), and it is he
who acts in the beginning of the next verse (‘and he slaughtered’). When
the information in this plus is taken into consideration, v. 28b in MT is no
longer unusual: the verb in that verse now becomes understandable,
since Elkanah had been introduced in v. 24 which had been lost by a
textual mishap. Besides, Elkanah was mentioned also in the reconstruc-
ted original text of the first words of v. 24 (see below).!7
When the text omitted by homoioteleuton is restored to its proper
place in vv. 24-25, MT is understandable, but not all problems are solved.
Hannah went to Shiloh together with her husband Elkanah and her son
Samuel, even though it is she who is the main actor at this stage of the
story. It is she who brings her vow to completion, and it is she who
presents her Song. But there are two elements that remain unclear in the
story in MT. Even though we now understand that it 1s Elkanah who
bows to the Lord in 1:28b, it is not clear why he should be singled out for
mention, thus omitting reference to his wife and son. It does not suffice
to point to the central place of men in worship. It is even more strange
that the story ends with 2:11, referring to Elkanah’s returning home and
Samuel’s serving the Lord. What happened to Hannah and why was she
not mentioned at the end of this episode in the same way she was in the
beginning and middle of the story? To this issue we shall return below.
The LXX presents a different picture of 1:28 and 2:11, partly shared
with 4QSam2. That the Greek translation does not reflect the translator’s
exegesis is demonstrated by the similar evidence of the Hebrew scroll,
even though some of the words found in 2:11 in the LXX appear in the
scroll at a different location, 1:28. The picture reflected in the LXX differs
from MT, since the statement of MT in 1:28 (see above and below) is
16 Pace Walters, “Hannah and Anna,” 403-404.
After the textual mishap in vv. 24-25 (homoioteleuton), ‘and he slaughtered’ (thus the
LXX and the reconstructed text of 4QSam®), referring to Elkanah, was not understandable
anymore and was made into a plural form in MT. On the other hand, according to
Wellhausen, Samuel, 41 and Driver, Samuel, 21 unmentioned ‘persons who slaughter’ are
the subject of the plural verb, reflecting a possibly original reading.
SONG OF HANNAH 439
lacking in the LXX, and in the similar statement in that translation in
2:11, it is Hannah who acts, not Elkanah. In fact, more or less the same
actions as are ascribed in MT to Elkanah are ascribed to Hannah in the
LXX and 4QSamé® (with internal differences): an action connected with
the Lord (prostration to the Lord in MT and 4QSam? and the entrusting
of Samuel to the service of the Lord in the LXX and 4QSam4) and
returning home to Ramah at the end of the action.
Furthermore, although similar actions are described in the different
versions, it should be noted that according to the LXX and 4QSam, it is
Hannah who leaves Samuel behind in the temple for the temple service.
The entrusting of Samuel to the temple is not mentioned explicitly in MT,
although it is implied by 2:11b.
In sum, the relation between the texts is now clear: The main
difference between MT on the one hand and the LXX and 4QSam/ on the
other is that in certain episodes in the latter two texts, Hannah acts as the
main character, while in MT there are two main characters, Hannah and
Elkanah. These two versions of the story are not parallel original
versions, as suggested by Walters.!8 Rather, they are genetically related.
Either MT ascribed actions to Elkanah which in an earlier version had
been ascribed to Hannah, or vice versa.!? We opt for the first possibility ?°
because of the contextual difficulties in 1:28 and 2:11 in MT. Especially
difficult is 2:11 in MT: since at this juncture Hannah should be
18 There is no room for an extensive discussion of the abstract concepts behind Walters’s
views, which center around the question of the original text and the transmission of the
biblical books. For the latest formulation of my own views, see my TCHB, 164-180. In
Walters’s detailed description of the differences between the versions, the Qumran
evidence is not sufficiently taken into consideration, and in our view Walters does not
distinguish between the translator’s exegesis (which is not relevant in the present context)
and his deviations based on a reconstructed Vorlage differing from MT. The translator’s
problems when encountering difficult words are also not taken into consideration. For
example, Walters tabulated major differences between the two texts in v. 6 in parallel
columns, but a great part of these differences derives from the translator’s understanding of
his Vorlage. According to Walters (p. 394), ‘M’s story describes Hannah’s difficult situation
objectively ... But B’s <that is MS B of the LXX> story, containing no provocatrice, describes
the situation entirely in terms of Anna’s subjective responses: she suffered thlipsis and
athymuia, distress and depression.’ However, the difference between the two texts derives
partly from the translator’s misunderstanding of an7x, ‘co-wife,’ which he took as ‘her
distress.’ In the wake of analyses of this type, Walters concludes (p. 392): ‘Both by its
omission of Hannah’s deferential reply and by the character of her first direct speech, M
portrays Hannah more positively than B, giving to her person—both words and actions—a
more substantive importance.’ This characterization is questionable.
9 Thus Driver, Samuel, 22: ‘LXX ... an addition to MT, which looks like a various
recension of the words not expressed by them in 1, Der
20 Also Wellhausen, Samuel, 42 prefers the reading of the LXX, arguing that it would not
make sense for the Greek text to suppress the involvement of Elkanah, which has been
mentioned in detail in v. 24.
440 CHAPTER TWENTY-NINE
considered the main person, it is strange that nothing is said in this verse
about her movements. It is thus likely that the statements about Elkanah
replaced the earlier story. The earlier version (the LXX and 4QSam)
ascribed certain actions to Hannah, which have been removed in MT,
while similar statements were inserted about Elkanah. The impression is
created that MT did not wish to assign these actions to Hannah since she
was a woman, and it would not be appropriate that a woman should
play such a central role in the story.
This assumption is supported by two other verses in the story in
which a similar tendency of suppressing Hannah’s actions is visible in
MT:
(i) 1 Sam 1:23 MT 1127 XR 7 OP? 7X (= T V)
May the Lord fulfill His word.
4QSam@ = 78n kava aly ap qx] (= LXX)
[May the Lo]rd [fulfill] that which comes
out of your mouth.
The two formulations differ in content, since MT refers to the word of the
Lord, while 4QSam? = LXX refer to Hannah's vow. It is difficult to decide
between these two readings, and, therefore, both readings could be
original. On the basis of Num 30:3, which deals with vows, both readings
are equally possible in this context: Mwy? PDD XY 72D 127 2M N?, “he
must not break his word, but must carry out all that has crossed his lips
(literally: came out of his mouth).’ According to a different train of
thought, however, only one reading was original, while the other one
reflects a later correction. It is possible that the reading of MT reflects a
correction of the text of 40Sam? = LXX: the mentioning of the ‘word’ of
God in MT reflects more reverence toward God than the vow of a mere
mortal, Hannah.*! MT thus did not mention Hannah’s vow explicitly.
(i)1Sam1:25 MT “3M 28 WIT NN WIM TA AX wow.
LXX: Oy Ox NIN ON TIN RIM AT NX onw)
Kal €odactev Tov pdcxov Kal Tpoonyayev Avva
1) LA\TNp Tot tatSaplou trp6¢g HAL
According to MT, unmentioned persons bring the boy to Eli, while according to
the LXX, ‘Hannah, the mother of the boy,’ comes to Eli. The connection in MT is
strange, since v. 26, referring to Hannah’s conversation with Eli, is not connected
with the previous verse, while the LXX presents a more logical context. It is not
21 it is also possible that the reading of MT was corrected in 4QSam@= LXX, possibly
since the ‘word’ of the Lord was not mentioned earlier in the text. For a comparative
analysis of these readings, see Tov, TCHB, 176.
SONG OF HANNAH 44]
impossible that the original wording was changed in MT in order to avoid
mentioning another one of Hannah’s actions.
c. 1 Sam 2:1
The Song of Hannah in MT starts with apxm ain 3725nm, while the LXX,
which does not represent the first two words, merely reads: kal eltev.
The evidence of 4QSam¢ is not clear because of the fragmentary status of
this text, in which the first two words could have occurred in the lacuna,
but could also have been lacking. Most probably in the earlier text form
(that is, the LXX), Hannah’s Song was not referred to as a ‘prayer.’??
Probably the prayer element was added in the introduction to the Song,
on the basis of 1 Sam 1:26, since the Song is not written in the form of a
prayer. Rather, it is a song of thanksgiving of the individual, referring to
a personal calamity experienced by the psalmist, and most likely the
Song was added to the story secondarily. The textual evidence thus
testifies to two stages in the editing of the Song of Hannah.
d. 1 Sam 2:2
LXX 4QSama MT
TTD WIT? PR [M] 7D wITP PR TTD WTP PR a
WAPND PVE PRI [ |
qN22 *wWITP PR A hi2d wp | Aa: PR b
WAND WE PN WDND WA PN c
Some notes on the reconstruction of the LXX:?8
ws | At first sight, it seems as if 8lkatos in the second stich of the LXX points
to a Vorlage different from MT 7x, but the Greek rendering should be seen in the
light of the different renderings of x elsewhere in the LXX.*4 The various Greek
translation equivalents of this word reflect an avoidance of a literal rendering of
wy as a designation of God.” Such a tendency may also be assumed in this verse.
It is thus methodologically questionable to reconstruct °7¥ here and to assume a
graphic similarity between the two Hebrew words.
22 Cf. the laconic statement of Smith, Samuel, 15: GB has simply «at e{tev, which is
enough.’
Problematic elements in the reconstruction of the LXX and the Qumran scroll have
been indicated with an asterisk.
24 Thus also Stoebe, Samuel, 101.
25 CF. A. Wiegand, “Der Gottesname 1X und seine Deutung in dem Sinne Bildner und
Schopfer in den alten judischen Literatur,” ZAW 10 (1890) 85-96; A. Passioni dell’ Acqua,
“La metafora biblica di Dio Roccia e la sua soppressione nelle antiche versioni,” Ephem.
Liturgicae 91 (1977) 417-453.
44? CHAPTER TWENTY-NINE
wip ]dytos in the third stich of the LXX may reflect w17?, which could also
have been contained in the lacuna in 4QSam4, but it may also reflect a free
addition of the LXX to the otherwise unusual*® phrase N72 PR °D.
The differences among the three witnesses pertain to major details, but
not all of them can be analyzed because of the uncertainty of the
reconstruction of the Vorlage of the LXX and the fragmentary status of
the Qumran scroll. However, at least this is clear:
1. A calculation of the length of the lines in 4QSam? makes it likely
that the scroll contained additional text, probably a stich, after v7? [XR
[m]iD. This stich has been reconstructed by some scholars?’ as ?°7¥ [°X)
wenoND on the basis of the LXX. However, the reconstruction of S{kato¢ as
2°7x% is far from certain (see above), and furthermore it is not at all clear
which text would have been included in the lacuna in the scroll.
2. The internal sequence of at least two of the witnesses differs. If the
three stichs in MT are taken as point of departure for the description and
are therefore denoted as abc, the sequence of the LXX is represented as
acb; if the LXX reflects different readings, as presented above, this
sequence should be represented as ac’b’. The sequence of the Qumran
fragment is represented as alx]bc, in which x represents either a’, b’, orc’,
or a different stich (d). According to this description, the three witnesses
reflect different versions (editions) of the biblical verse.
It is hard to know which arrangement is preferable. The difficulties
inherent in the sequence of MT have often been pointed out, since stich b
in MT starts with >2, even though it does not explain the previous one.
e. 1 Sam 2:8
MT a. 7X OP NNW 27 7DVN OPN
b. Ben? 7129 NOD OI aY win?
CIN O12 DUP PIN Pea in 2
a. He raises the poor from the dust, lifts up the needy from the
dunghill,
b. seating them with nobles, granting them seats of honor.
- Comparable are only 2 Sam 7:22 and 1 Chr 17:20 where the Hebrew (and Greek) text
has an element describing the preposition: N71 O78 JX) 1VD PR *>.
27 Thus Cross, “New Qumran Fragment.,” 26 and Ulrich, Samuel, 121. A.L. Warren
argues that the plus in the LXX, and independently the plus in 4QSam%, reflect liturgical
expansions: “A Trisagion Inserted in the 4QSam® version of the Song of Hannah, 1 Sam.
2:1-10,” JJS 45 (1994) 278-285 (... LXX has been subject to liturgical adaptation for the
autumn Rosh Hashanah festival, probably on the basis that Elkanah’s annual pilgrimages
were also at this time of the year.’; p. 281). This article elaborates on H.St.J. Thackeray, “The
Song of Hannah and Other Lessons and Psalms for the Jewish New Year’s Day,” JT5 16
(1914) 177-204. See further below, n. 51.
SONG OF HANNAH 443
c. For the pillars of the earth are the Lord’s, He has set the world
upon them.?8
The three textual witnesses for the third segment of this verse run as
follows:
MT Zan onP2y nw prqN yp¥a mM? “dD
4Q0Sam? = 9 Jan ont>y [nw prs "pen a> 7D
LXX >
This third part of v. 8 is not represented in the LXX. V. 8c was probably
lacking in the Vorlage of that translation, and was added in a different
and later edition,*? represented by MT and 4QSam?. There are no literal
parallels for this verse elsewhere in the Bible,?9 and it would probably
have originated within the tradition of the Song of Hannah during one of
the stages of its growth. It represents a causal clause, supposedly
explaining the previous ones, although in actuality it does not provide an
explanation or background for them.
The background of v. 8c should be understood in the light of its
relation with the surrounding verses. The first two segments of v. 8, as
well as the next verse, 9, deal with the fate of individuals, while v. 8c, the
added clause of MT and 4QSam4, deals with God’s universal powers.
What all verses in this context have in common is that they stress God’s
power in determining the fate of the individual. But, while vv. 4-9 (with
the exception of 8c) deal with God’s ability to determine the fate of
individuals, 8c mentions God’s cosmic powers.?! Verse 8c is phrased as
an explanation of the preceding verses, but since it mentions God’s
universal power, it fails to do so. When faced with texts which either
contain (MT and 4QSam#) or lack v. 8c (LXX), one should probably
consider the text that does contain v. 8c as secondary. The universal
power of God is mentioned again in v. 10, but in that verse the
description of this type of cosmic power fits the description of God’s
overpowering his enemies. The juxtaposition of a description of the
personal fate of individuals and God’s greatness in the universe is found
also in Psalm 113, which in many ways resembles the Song of Hannah,
28 The translation of this verse, as well as all other ones, follows NJPST.
29 Thus already Wellhausen, Samuel, 43 and Smith, Samuel, 16. Neither scholar uses a
term such as ‘edition’ or ‘interpolation.’
30 For the idea cf. Ps 75:3-4, according to which God's giving judgment equitably is
paralleled with His keeping the pillars of the earth firm. Other parallels are mentioned by
Tournay, “Cantique,” 563.
This was recognized by Dhorme,”Cantique,” 391 (cf. n. 3).
444 CHAPTER TWENTY-NINE
but that fact cannot be used as an argument in favor of the originality of
v. 8c of MT.
The presentation of 8c in the different versions is somehow related to
v. 9a. That verse, too, is not represented in the LXX, which presents a
different verse in its stead. However, vv. 8c and 9a should be dealt with
separately because of their different structure. V. 8c supposedly explains
the preceding clauses 8a-b, while 9a contains a new idea, for which 9b
forms an explanation.°2
The LXX stands alone in the non representation of v. 8c, against MT
and 4Q0Sam4, in which it is found. Likewise, the LXX is alone in not
representing v. 9a. In other deviations from MT, the LXX is joined by
4Q0Sam4, as shown below.
1 Sam 2:9
MT a. WT WN ovwT Anw? reon 77
He guards the steps of His faithful, but the wicked
perish in darkness,
a’. >
b. WN 727 132 X27
for not by strength shall man prevail.
LXX a. >
a’. PTR maw /w I TW? 77 7N3
b. wN TW MDI Xx? 7D
BLB0US ELXNV TU ELVYOWEVW Kal EvAdyNOEV ET Stkalou STL
OUK €V Loxtt Suvatos dvip
He gives the vower his vow and blesses the years of
the just,
for not by strength shall man prevail.
hat qwna oywn Anew rto]n 77
[He guards] the way of [His] fa[ithful, but the wicked
perish in darkness. |
a’. pax myw/r3]w FAI A[77] 771 ys
He gives [the vow]er his vow and blesses the y[ears of
the just],
Be T20R ay AoE? 5]
[for not by strength shall man prevail (?)].
4Q0Sam4
-
32 tt is unfortunate that de Boer in BHS did not provide the text itself of the LXX, for the
note ‘G alit’ underestimates the importance of that evidence. It is somewhat misleading
that the evidence of 4QSam? relating to v. 9a is listed in a note to v. 8c.
SONG OF HANNAH 445
The earlier text of vv. 8-9 probably consisted of only 8ab and 9b. This text
was revised in different ways in MT and the Vorlage of the LXX. 4QSam4
represents a hybrid version.
V. 9a of MT and 4QSam? °S is not represented in the LXX. In a way,
the idea of v. 9a (in MT: 17> Qwna o-ywT Taw? TON 7719; 4OSam? presents
a slightly different formulation) fits the Song of Hannah. That Song
mentions in vv. 4-5 unexpected changes for the better and the worse in
the fate of individuals. Likewise in vv. 6-8 the Song mentions God’s
power to change the personal fate of individuals. The implication of
these two groups of verses is that the unexpected change in condition
(for example, the strong whose power fails, in v. 4a) is due to God, who
can bring about these changes, just as He can make the poor rich, and the
rich poor (v. 7). The descriptions in vv. 4-8 serve as examples of God’s
power mentioned in v. 3, and they are in line with the general praise of
God in vv. 1-3. This idea of God’s almighty power underlies a]] textual
traditions of the Song and is also behind v. 9b (w°x 7237 732 Xx? *D), but is
made more specific in MT and 4QSam# in v. 9a. The implication of vv. 1-
3 and 6-8 for vv. 4-5 is that God’s power is behind the changes in the fate
of the individual mentioned in those verses as well. Vv. 4-8 are therefore
understandable as they are. The main idea of these verses is that changes
will occur if God wants them to occur. However, in two textual
traditions, MT and 4QSam, this idea has been elaborated upon and been
made more specific. One might say that the original ideas have been
given a theological slant. The presumably earlier stage in which the
original idea of the Song was represented has not been preserved. The
existence of such an earlier stage reflecting a shorter text is reconstructed,
although it is supported by the LXX. That support is only partial, since
the LXX itself has expanded the originally shorter text. It is suggested
here that the originally short version of the Song, lacking vv. 8c and 9a,
was expanded in one direction in MT and in another one in the Vorlage of
the LXX. 4QSam? contains a hybrid text.
It is suggested here that the text common to MT and 4QSam? in v. 9a,
and lacking in the LXX, represents a theological elaboration on the main
theme of the Song of Hannah. That addition to the original text
reinterprets the examples of the changes in the fate of the individual
given in vv. 4-8 in a certain way. According to this reinterpretation, the
sudden changes described in those verses do not exemplify the strength
of God, but the power of loyalty to God. It is the person who is loyal to
33 Only the first word of v. 9a has been preserved in 4QSam@; the remainder would have
been contained in the lacuna.
For the reading of 4QSam® cf. Prov 2:8.
446 CHAPTER TWENTY-NINE
God who will experience a change to a good condition, and it is the
wicked (that is, the ones who are not loyal to God) who will experience a
change to a bad condition. This reinterpretation found in MT and
4QSamé of v. 9a was probably added to the Song.°° It was the intention
of the person who added v. 9a that the contents of this verse would be
applied to vv. 4-8.
A different revision*© of the earlier shorter text is found in the LXX
which (together with 4QSam4) contains a completely different text, v. 9a’.
On the special status of the Qumran scroll, see below.
The verse which is found in the LXX and 4QSam4 (47275 47117] 473 7M
[P78 miw/71]w) and not in MT, is secondary because of its content and
position. After mentioning the various categories of change from evil toa
good condition and from a good condition to a bad one, and after
mentioning God’s power in bringing about these changes,?” it is
somewhat anticlimactic to mention in v. 9a’ God’s granting the vow to
the person who vows. God’s power is reflected in so many categories
that the granting of the vow“® seems to be a mere detail presented as an
afterthought. Since v. 9a’ is not found in MT, it may be suspected as
secondary, since it reflects the special situation of Hannah. This verse
clearly reflects an attempt to accommodate the Song more closely to
Hannab’s situation.*?
The second phrase of the LXX, kal evAdynoev Et dStkalou = 4QSam4
[j°7x miw/>1]v 77271, probably does not refer to the righteous in general,
but mentions them only in conjunction with the person who makes a
vow. The mention of the righteous in the LXX = 4QSam? thus runs
parallel to the mention of the persons who are loyal to God in v. 9a in
MT. The phrase of the LXX may be taken to imply that the persons who
witness a change in their personal fate, as mentioned in vv. 4-5, are the
righteous. In that case, the reinterpretation reflected in the LXX and
4QSam* runs parallel to that of MT.
3 For the understanding of the background of this verse, it is important to note that the
specific use of ywn, ‘wicked,’ as describing persons who are disloyal to God, occurs mainly
in Ezekiel, Psalms and the Wisdom literature.
36 Thus already Stoebe, Samuel, 102 with reference to the LXX (’... Rezension, die noch
starker die Situation berucksichtigt’).
3? This analysis is based only on the text of the LXX and disregards the pluses of MT.
The exact phrase of God’s ‘granting the vow to the person who vows’ is not known
from other verses, but is not intrinsically difficult.
? Driver, Samuel, 26. According to Wellhausen, Samuel, 42, the attempt is not successful,
since the Song presents God as granting more than his worshipers expect, while according
to v. 9a' God fulfills the wishes of the worshipers exactly.
SONG OF HANNAH 447
V. 9b wx Jay D3 x? 7D 49 is common to the MT and LXX and probably to
4QSam?, and its meaning in the different contexts of these witnesses
needs to be discussed next. In MT this sentence connects well with the
two preceding ones.*! Physical force does not give strength to people.
The idea of this stich could continue v. 9a, according to which the
righteous as well as the wicked will be judged according to their loyalty
to God; physical power (that is, of the wicked) will not help them. But
within v. 9 it appears to be an afterthought, since the main idea was
already expressed by v. 9a. There is no good connection between this
stich and its context in the LXX and 4QSam/. In these two sources the
third stich, mentioning the ineffectiveness of physical power, should
explain the two preceding stichs. In the words of the LXX: (a') &80bs
ELXNY TU EvyoLevw Kal ELAdynoEV ET Stkalov (b) Sti ok Ev Laytt Svvatds
dup. In our view, there is no necessary connection between the ideas of
a’ and b. This lack of connection may indicate that 9a‘ of the LXX and
4Q0Sam, lacking in MT, contains an editorial insertion into the text.
The contextual appropriateness of 9b in the reconstructed original text of
the Song of Hannah needs to be discussed next. In the reconstructed text,
which lacked v. 8c and 9a of MT, 9b immediately followed upon 8b. The
reference to the ineffectiveness of physical power in v. 9b connects well
with v. 8b.
If the above analysis is correct, MT and LXX = 4QSam/ reflect two
different and independent reinterpretations of the main ideas of an
earlier form of the Song of Hannah.
When the different forms of this verse are compared, we are
confronted with three different versions, which may be represented
schematically as
MT ab
4QSam? aa'c
LXX a'b
In this web of relations between the versions, 4QSam? holds a peculiar
position. The text of that scroll is closely related with the LXX against
MT, since it contains the secondary verse about God’s ‘granting the vow
to the person who vows’ (9a'). However, the scroll also agrees with MT
against the LXX in preserving another secondary addition, viz., v. 9a of
MT. According to our analysis, the additions of the MT and LXX present
40 For the idea and words, cf. Zech 4:6; Job 21:7.
1 According to Tournay, “Cantique,” 564, v. 9b is connected with the next verse (10) in
spite of the verse division of MT. Stoebe, Samuel, 102, following others, considers v. 9b to be
a secondary addition.
448 CHAPTER TWENTY-NINE
two different types of reinterpretation and contextual adaptation of the
Song of Hannah, so that their juxtaposition in 4QSam/ is very peculiar. In
our view, since 9a of MT and 9a’ of the LXX are contextually secondary,
their combination in 4QSamé4 should be considered secondary as well.
The juxtaposition probably represents a textual mishap*? or a scribe’s
wish to present both versions. Ulrich and McCarter’ suggest a different
type of solution, according to which the text of 4QSamé@ reflects an
original text from which the other two texts developed because of a
textual mishap, named haplography by these scholars.44 However such a
presumed development does not explain the text of the LXX. Besides, the
methodological argument mentioned above is even stronger: the
juxtaposition in 4QSam#? of two intrinsically secondary verses should be
regarded as non-original.
On the basis of the aforementioned considerations relating to three
textual witnesses, the following stages in the development of v. 9 are
reconstructed:
stage 1: 9b w°x 722° N32 XN? °D (all witnesses; connected with 8a-b)
stage 2a: addition in MT of 9a wt Jwna o-ywn Tw YON *717 before
9b
stage 2b: addition in the Vorlage of the LXX of 9a‘ Jr 479 771: JM}
>-Tx nuw/ vw before 9b
stage 3: combination of texts reflecting stages 2a and 2b in 4QSam?:
a. [oT Jwna oywn Anw ITO]N 7AM
als (Prem /y aa ae a] m
bb: 2 soa bet gs ae a)
There is room in this reconstruction*® for v. 9b, but it is unclear whether
this sentence was included in the scroll. It would have appeared at the
end of a line, but instead, the remainder of the line could also have been
empty (‘open section’). The latter assumption is unlikely as this would be
the only paragraph marker in the Song of Hannah.
g. 1 Sam 2:10
1 Sam 2:10 in the LXX and 4QSamé? differs completely from MT. Both
texts add a long section after the first segment which they have in
common with MT (ja inn? 717 in MT and a slightly different form in
42 Thus Barthélemy, CT 1, 145.
43 Ulrich, Samuel, 120 and McCarter, Samuel, 70.
44 Tt is not clear what kind of haplography one should have in mind. The only
haplography (or rather homoioarcton or homoioteleuton) which comes to mind is between
mm") ~D at the beginning of v. 8c and x? “> at the beginning of v. 9b.
© This reconstruction does not follow the layout of 4QSam?.
SONG OF HANNAH 449
4QSam? and LXX). In this case MT contains the earlier form, while the
LXX reflects a long exegetical plus which is in the nature of an
afterthought. The contents of the plus in the Qumran scroll, preserved
fragmentarily, cannot be identified easily, but it is found in the same
position as the plus in the LXX and possibly has one phrase in common
with the LXX (wiq]j? *n; see below). The addition of the LXX, which is
presented here together with the surrounding verses, runs as follows in
Greek and in its reconstructed Vorlage:*®
1Sam 2:10 MT _ reconstructed Vorlage LXX
wp wn? 7 Vann *nne4? ma Kbptos dabevi Trottier
avTLBtKov abtov
wit mim48 Kiiptos d&yLos
WANN. ODN 2AM 2X ph KavydaOw 6 ppdvipos ev
TH dpovyoet avTov
NWA WA. PAN ON Kal ph KavxdoOw 6 Suvatds
EV TH Suvdptet avTtov
Twyla wy 2am box Kal ut) Kavydobw 6 TrotOLOs
EV TH TAOUTW AUTO
22nd 220M NkID OND AAN F ev tottus Kavydobw 6
KQUXWLEVOS
mam nk yt 2Dwn ouvlewv Kal yLvwoketv TOV
KUPLOV
YUX3* APTS) odDvn WwY) Kal trovetv Kplua Kal StKaLO-
OUND EV LEGW TTS YS
PAX "OR PR TT PUN 7OON PT TI abLTds KpLVveT Akpa yhs
D2n wy yn *D9n7 WIN Kal &lSworv Loxyby Tois
Bao.re toy Hav
mwn py ow mw APO Kal VbwoEL KEpAasS YPLOTOU
auTot
With a few differences the plus of the LXX reflects the MT of Jer 9:22-23
which is presented below together with the Greek text of 1 Sam 2:10 and
its reconstructed Vorlage:
MT of Jer 9:22-23 reconstructed Vorlage LXX of Samuel
of LXX-Samuel
‘naan oon OA NX inpana oon BAM ON Ut Kavxydobw 4 dpdvipos Ev TH
ppovyoer abtov
46 Stars indicate especially problematic reconstructions.
47 Thus also 4QSam?: nn.
8 See the discussion below of a possible equivalent of this Greek plus in 4QSam‘®.
450 CHAPTER TWENTY-NINE
NI WIw WaM PR yn71932 Way* PoAN? DN Kal Lt Kavxydo@w 6 BSuvaTds év
TH Suvduer avtoi
Mwy ywy An OR Neva vey Dam Dw Kal pt Kavxdo8w 6 TrAObGLaS Ev
TS TAOUTW abTod
boanon Dorn nara OR °D Doan 2AM nRID OND «GAN F Ev ToUTW kavxdobw 6
KAVXWLEVOS
af RIN OSU (717)? nx vr down ouvlewv kal ywwusake rv Tov
KUPLOV
PARA TPT Dw... TWY PARA APTA vow MWY Kal Tovetv kplya kal
Sikaroauvny év péow THs YTS
The differences between the MT of Jeremiah and the reconstructed
Vorlage of the LXX of Samuel are indicated with a larger font. The
additional words of the MT of Jeremiah are indicated by ellipsis dots.
The added verses in the Greek translation of Samuel have not been
transferred from the Greek translation of Jeremiah since they differ in
several details, as indicated below, reflecting either different translation
equivalents or differences in Hebrew Vorlagen.*?
LXX of Jeremiah LXX of Samuel
Ln Kavxdoduw 6 codds Ev TH) codla LT) Kavxdodw 6 dpdviptos év TH
avTov ppovyioe. avTov
49
Tordv in the LXX of Jeremiah reflects the participle nwy also found in MT, while kal
trocetv of the LXX of Samuel points to an infinitive reflecting a different reading of the same
consonants, viz., 1vy. Likewise Tdv kuptov of Samuel reflects a Hebrew reading different
from the MT of Samuel, probably ’* nx (the abbreviated tetragrammaton), which is closely
related to "mx of MT of that book. I.L. Seeligmann, Studies in Biblical Literature (Hebrew;
eds. A. Hurvitz and others; Jerusalem 1992) 325-326 rightly considers the Hebrew reading
behind the Greek translation of Samuel to represent the original meaning of the context in
Jeremiah, and not MT ad locum. The Samuel text speaks about two desirable actions for
mankind, knowing God and acting with justice, while the Jeremiah text speaks of man’s
knowing God and God's acting with justice. These two readings have to be evaluated in the
light of the meaning of the context. That context creates a certain opposition between the
boasting of men about certain qualities and possessions on the one hand and religious
virtues on the other. Within that framework the clearly defined opposition between the
actions and views of humans, as in the addition in Samuel, is more natural than in
Jeremiah. It is, morever, unusual that the text in Jeremiah stresses in a somewhat
tautological formulation the fact that God acts with justice. ‘Is it appropriate that the God of
the universe claims that He acts with kindness, justice and equity in the world, for in these
He delights’ (Seeligmann, ibid., 326). According to Seeligmann God is depicted here as an
arbitrary ruler, doing only what He wants. It so happens that the formulation preserved in
Samuel is contextually more appropriate to Jeremiah and also reflects the terminology of
that book better (cf. Jer 22:15-16). Probably the text which is now preserved only in Samuel
once served as the original text of Jeremiah; it was slightly corrupted by a misreading of an
abbreviated tetragrammaton, and this misreading caused a series of contextual adaptations
in the text of Jeremiah.
SONG OF HANNAH 451
Kal Un Kavyao8w 6 layupds Ev TH Kal Un Kavydotw 6 duvatos év TH
loyu avtTow Suvdpet avtow
Kal pt) KavydoOw O TAOUGLOS EV Tui Kal Ly) Kavydo0w 6 TAOUGLOS EV TU
TAOUTW avTot TAOUTO) AUTOT
aX 1 Ev ToUTW Kauxdabw 6 adX TEV TOUTW KavyYaG8u O
KAUXWUEVOS KQUXWLEVOS
OuvLEeLV Kal YLUsOKELV OTL Eya Ely OUVLELV Kal YLYWOKELVY TOV KUPLOV
KUPLOS
TrOLwV EA€os Kal Kplua Kal Kal Trotetv Kplpa Kal 8tKaLoaUUNY éV
SLKALOGUUMD ETL THS YTS péaw HS Ys
Since the plus of the LXX in Samuel did not derive from the Greek
translation of Jeremiah, it was most probably based on a Hebrew plus,
such as reconstructed above. This addition has contextual relevance,
though not necessarily in its present place, where it occurs after the first
clause in the translation of v. 10, representing 12-77 inne AW of MT.
However, the plus of the LXX is actually connected with 1214 122 x? 7°
wx, that is, the last words of v. 9. It also refers back to v. 3 27Nn 127N 3x
mia 4722, which in the Greek translation is represented by the same
Greek verb representing ??0m ?x, namely ut Kavydo@e .... (influence on
the Greek level, however, has been discarded above).
On the one hand, it is hard to imagine a running Hebrew text which
would be worded like the Hebrew text reconstructed from the LXX
because the addition based on Jer 9:22-23 is located inappropriately
between two phrases of the Hebrew which are closely related to each
other, viz., after 13°7n nn? TAI" and before ayy oO-nwa iv. On the other
hand, 4QSam¢? has an equally long addition at exactly the same point, so
that this addition or a similar one indeed formed part of a Hebrew text.
This assumption may be strengthened by two elements of the addition,
which did not derive from the Jeremiah context, but which are also
found in 4QSam?: The words immediately preceding the quotation from
Jeremiah in v. 10, kUptos dytos, probably represent wip 17°, and they
stand exactly at the same place as a plus in 4QSamé@ in 1. 29: ]j> 7,
probably to be reconstructed as w17])? 7».
The plus of 4QSam? found at exactly at the same point as the long
plus in the LXX cannot be identified, due to its fragmentary nature. In
whatever way the last letter of the one complete word in the first line of
the plus is reconstructed, it is difficult to know what it means.°9 In any
50 Lewis, “Hannah,” 43 suggests that 1/4nw3, or in his reading *n?wa, is derived from
the root 07v, signifying ‘paying one’s vows,’ for which cf. the vow mentioned earlier in the
context. Por the reconstruction of the second line, cf. Prov 2:8. A completely different
reading of that line, if correct, would bring the addition of the scroll into close contact with
452 CHAPTER TWENTY-NINE
event, the few preserved letters cannot be correlated to the plus of the
LXX which is based on Jeremiah 9:
[ JA/yo2w3 OPool J
[ Taw? Po} 6737 ]
| }*n of
Little can be said about the location of the two additions in the LXX and
4QSam?. The addition of the LXX seems to be out of place. The small
addition kuptos dytos is not at all connected with its context. Rather, it
harks back to v. 2, where the same words occur. The long addition is
inappropriately located between two phrases of the Hebrew (see above).
The flow of the text is logical in MT, but it is not in the LXX. While the
assumption of marginal notations must be reduced to a minimum (see
Tov, “Glosses’”’*), it seems that this is a case of such a marginal note or
possibly two notes, originally placed in the margin and afterwards
wrongly inserted into the text itself. This marginal note is not in the
nature of a gloss or an interpolation, but rather a reader’s remark.?!
The short text of MT is thus original in this place, while the plus(es)
reflected in the LXX and 4QSam/? represent (a) contextual afterthought(s).
3. Conclusions
The three versions of the Song of Hannah differ in many small details,
not discussed here, as well as in the several major details.
In its original form, the so-called Song of Hannah reflects a thanks-
giving hymn by an individual,°* referring to various situations in which
that individual was saved from a calamity. This psalm stresses the
special power which enables God to bring about changes, especially
changes to a good condition, or to a bad condition.» The major stress in
the LXX. Cross, whose reading is quoted by Lewis, “Hannah,” reads [?]9[1]m 2x, which
would be identical with the text of Jer 9:22 quoted above. However, while our own reading,
previously quoted in the scholarly literature by others, is problematic, the reading
suggested by Cross (if based on PAM 43.122) is even more problematic: Between the first
letter (read by Cross as an aleph) and the next letter, read as a lamed, there is a clear remnant
of another letter (read by us as a gimel). Further, the speck of ink which is read as a lamed in
2] [71] is positioned at the place where a he is expected, and if this were indeed a remnant
of a lamed, probably a greater part of that letter would have been preserved.
ol Thackeray suggested long ago that the addition in the LXX was based on Jer 9:22-23,
read as a short haphtarah together with the longer haphtarah of the Song of Hannah on the
New Year's day. See H.St.J. Thackeray, “The Song” (n. 25) esp. 190.
2 For a different characterization, see Willis, “Song” (see n. 3).
3 In this regard Hannah's Song is very similar to the Magnificat in Luke 1:46-55, both in
general ideas and in wording.
SONG OF HANNAH 453
this psalm is on the change to a good condition, and the psalmist thanks
and praises God for this change in his fate.
The main idea of the reconstructed original form of the Song has been
reinterpreted in two directions in the preserved texts, which therefore
constitute different editions of the Song and its narrative framework.
These changes are evidenced in the three main textual sources which
have been preserved, either in individual witnesses, or in groups of two,
without any consistency. Sometimes the change is evidenced in MT, and
sometimes in the LXX, and either one is sometimes joined by 4QSamé?.
The position of 4QSam? is thus rather peculiar, but in any event it is
through the evidence of this scroll that the real nature of the two other
texts comes to light. Each of these witnesses makes the Song of Hannah
more relevant to its context or to certain trends in biblical theology.
Furthermore, MT replaces the role of Hannah in the story of the third
visit to Shiloh with actions by Elkanah.
The ‘original’ form of the Song cannot be reconstructed, but
presumably it was shorter than the text now transmitted in the three
Witnesses. It is not impossible that that original form was even shorter
than implied by the above analysis, since v. 10, with its description of
God’s universal powers and the messianic ending, may have been added
to the original Song,9°4 even though this view is not supported by any
textual evidence.
The differences between the three sources may be summarized as
follows:
a. The Song of Hannah is located in two slightly different positions in
MT on the one hand and the LXX and 4QSam/? on the other.
b. The three texts present different concepts of the events occurring
before and after the Song (1:28; 2:11). The LXX and 4QSam? seem to
present a more original and internally consistent version. MT reflects a
revision which shifts to Elkanah a role which was originally ascribed to
Hannah.
c. The original form of 2:1 has been preserved in the shorter version of
the LXX. The edition of MT adapted the Song to the context by an
addition which makes the Song into a prayer.
d. 2:2 has been preserved in three different editorial forms. The
possible stages in the editing of this verse cannot be reconstructed.
e. 2:8c, lacking in the LXX, was added in MT and 4QSam, in order to
stress the universal power of God not mentioned in the earlier edition.
94 Thus O. Loretz, “Psalmenstudien (II),” Ugarit-Forschungen 5 (1973) 214. For a
discussion, see Tournay, “Cantique,” 565.
454 CHAPTER TWENTY-NINE
This universal power is also referred to in v. 10, where it suits the
context.
f. The reconstructed earlier text of 2:8-9 consisted of 8ab and 9b only.
This short text was interpreted in two different ways in MT and the
Vorlage of the LXX. 4QSam4 represents a hybrid version of both
reworkings and thus represents a secondary development.
2:9a, common to MT and 4QSamé and lacking in the LXX, represents a
theological reinterpretation of the main theme of the Song of Hannah.
According to that verse, the person who is loyal to God will witness a
change for the better, and the wicked (that is, the ones who are not loyal
to God) will witness a change for the worse.
The counterpart to 2:9a in the LXX, v. 9a’, reflects an attempt to
accommodate the Song more closely to Hannah’s position by adding a
reference to God’s granting the vow to the person who vows.
g. 2:10 in MT differs completely from the LXX and 4QSam‘?. The latter
two texts add a long plus after the first segment, which they have in
common with MT (12°79 inn? 717 in MT and a slightly different form in
4QSam? and the LXX). MT contains the earlier form, while the LXX
reflects a long exegetical remark which is in the nature of an
afterthought. This afterthought was inappropriately added between the
components of v. 10. The contents of the plus in the Qumran scroll have
been preserved only fragmentarily, but the scroll may have reflected the
same plus as the LXX.
The analysis of the aforementioned differences is relevant to the
history of the biblical text as a whole. One may view editorial differences
between the textual witnesses as deriving from several irreducible
pristine versions of the biblical text. This has been suggested by various
scholars, and has been spelled out in detail by Walters, “Hannah and
Anna” for the first chapters of Samuel. In our view, however, almost all
instances of variation can be explained as revisions of some kind of an
earlier text, so that in spite of the major differences between the textual
witnesses, an Urtext of some kind can still be assumed.’ The main
objective of this study is to show that the Song of Hannah circulated in
antiquity in different recensions. These data are instructive for our
understanding of one stage in the development of the text of the Bible, in
which different recensions were extant. At the same time, these
assumptions need not bring about a change in our thinking about the
55 The readings of 4QSam@ and the LXX in 2:10, however, are described as marginal
notes.
SONG OF HANNAH 455
earlier stages of the biblical text. There is still room for a reconstructed
original form, the nature of which needs to be formulated carefully.
VI. REVISIONS OF THE SEPTUAGINT
CHAPTER THIRTY
PAP. GIESSEN 13, 19, 22, 26: A REVISION OF THE SEPTUAGINT?
1. The texts
Greek fragments of Deuteronomy 24-29 found in Antinoopolis were
published in 1911 by P. Glaue and A. Rahlfs as “Fragmente einer
griechischen Ubersetzung des samaritanischen Pentateuchs.”! These
fragments have been of considerable interest to scholars, since until
that time the lost translation of the SP was known only from 43 (46)
marginal notes and quotations, whereas the newly found fragments
preserved a large running text of the ZapapeitiKov (hereafter: Xaj.).
The fragments have been aptly described by Glaue and Rahlfs
(hereafter: GI.-R.), and their identification as Zap has not been
challenged.* Not without hesitation the present presentation differs
from that of GI.-R. in the interpretation of some renditions and in their
classification, and in the overall understanding of the nature of the
fragments. The majority of GI.-R.’s penetrating remarks on the
individual words (as opposed to their general view of the fragments)
remain as valuable as they were in 1911. The present article suggests
that the Giessen papyri do not reflect the Zapapettikdv; likewise, the
Samaritan inscription found in a synagogue in Saloniki does not reflect
that version (see Tov, “Samaritan Inscription”*). Rather, these two
sources reflect revisions, possibly of Samaritan origin, of the OG.
The present re-edition of the Pap. Giessen (henceforth: P.G.)
incorporates some new readings and reconstructions. Since the fragments
1 NAWG, Phil.-hist. Kl. 1911, 167-200 = MSU 1,2 (Berlin 1909-1915) 31-64.
The earlier literature was mentioned by B.K. Waltke, Prolegomena to the Samaritan
Pentateuch, unpubl. diss., Harvard University (Cambridge, MA 1965) 65 ff. On the
‘Samaritan-Hebrew’ Vorlage of the papyri, see P. Kahle, TSK 88 (1915) 403 = Opera Minora
(Leiden 1956) 6. For the modern literature, see S. Noja, “The Samareitikon,” in: A.D. Crown
(ed.), The Samaritans (Tubingen 1989) 408-412. The most thorough analysis of what is
known of the Zayapertikdév is presented by R. Pummer, “The Samareitikon Revisited,” in:
E.A. Crown (ed.), New Samaritan Studies—Essays in Honour of G.D. Sixdenter (Sydney 1996)
381-455.
460 CHAPTER THIRTY
themselves have perished,? the present edition is based on the
photograph of the most extensive fragment as published by Gl.-R and
on G.-R’s transcription of the other fragments.
Pap. Giessen 19, 1.2 verso right#
Deut 24:15 [ev cot aya lo[tia]
16 ovk atroBavo[vuv |
TAL TATEPES
UTED ULWY A | Tw mpo0a] Dt 24:20
ov8E ULOL UTED 5: NA[vTw Tw opda |
TATEPWY aTro v[w Kal TH xnpa]
[O]avouvTat. eo[tat]
[a]lvnp tétats a Ot[av tpvynons] 21
[LJaptiats atro To[v aptredwva |
17 @Oavitat’ ovK EK 10 ov [ouK etrava|
KNLVELS KPLOL To[vynoLo auTov |
TpoonNAUTOV Tw [TpoonAUTW|]
opdavov. Kat Tw [opdavw Kar]
XT|Pas OUK € ™ x[npa cotat|
ve XupLaons to: al ]
[ups JatLov if |
18 [kat] pyno8[no}y
[oT SJovAos n[oba]
[ev avy|uTTH n[oba]
[ J] 20
ee aa
ca. 4 lines
Goll TOLELY Tov |
Ao[yov TouTov |
19 Ota[v Beprons]
tov §[eptopov | 25
Tov ay[pou cov |
K[a |. etr[tAa@n|
[S]pay[ya ev Tw
3 CE HG. Gundel, Libri 6 (1955) 50-51; idem, “Die literarischen Papyri in der Giessener
Universitatsbibliothek,” Kurzbericht aus den Giessener Papyrus-Sammlungen 12 (1962) 20; id.,
“Papyri bibliothecae universitatis Gissensis,” ibid., 27 (1968) 8 ff.
4 The reconstructions and readings included in the apparatus sometimes differ from
those of GI.-R, mainly as a result of our contention that the P. Giessen are part of the LXX
tradition.
PAP. GIESSEN 461
I 20 [eduTp |wa[ato|]?
[25 Cf. Aquila ad loc. and Lev 23:10, 22 LXX.
I] 4-7 Our reconstruction. L. 7 is short because the next line
started with a new paragraph. Cf. fragm. 19,1 verso left,
line 15.
i The reconstruction of autov is to be preferred over otLow
cou (cf. MT ronx) because of spatial considerations.
1112-14 Contrast GI.-R tw [tpoondv](13)Tw.
1115-16 [ac wvnoGrnon](16) o[tt ovkeTHs |?
Pap. Giessen 19, 1.2 verso left
1 oe€tat To ov[o| 6
Wa autou €€ tna
eav 6€ un Bou
Deut 25:3 [paottywloalt] AnO7 o avnp 7
[ v 5 daBiy Thy yu
[ |tAn vatKa Tou aden
[ynv ToAAnv]. pou AUTOU
lacxnuovnge |t oa avaBnoeTat
[SeAdos Golu € TN yuvn Tova
[vavtiov ao jou 10 Sedov auTou
4 [ nas ELS THV Oupav
[ Im TIPOS TOUS OU
[ ] VETOU’ KGL EL
[ Jo TH oukK €Bou
5 ka 15 dn6n o ek Tlev]
[ }to Qepou po[u a]
[ is [vlaotpolat Tw]
[a]SeAb[w av]
tolv ovowa €V |
20 inlA oux née |
Mncev |
[ Kau]
[kakeoouot |v au 8
[Tov ot Gu|vETOL
[ms TroA]ews av]
25 [tov Kat] Aadn
[covotv | auTw
[kat otals ert
462 CHAPTER THIRTY
Ly [ynv Tony Kat?
[11 LXX: ou @LUWOELS
ibis [eEakerdOn loetat.
1115-16 GIL-R.: [te ]Oepou.
II 20-21 Gl.-R.: [eBov]A[nén].
If 21-22 GI.-R.: [o ek we 8epov pov] = ~n2° (sic LXX, SP!4F8); a
reconstruction ouvotkyoat auTy (sic LXX 25:5) = MT 727
would be as fitting.
IT 23-27 = [exadeoa]v and [e]AaAn[cav] (both more literal renditions of
MT than the present reconstructions) are not impossible, but
in that case the reconstruction of line 27 would be a little too
short. Moreover, avaBnoetat in line 8 is not a literal reflec-
tion of MT either.
Pap. Giessen 19, 1.2 recto right; 19, 3.4 recto?
Deut 27:4 ...eyw evTe|[AAo|
Hat ULV on[pe]
pov ev ap(?)yap|t]
Cu’ Kat xpto[ts |
QUTOUS XPLOTN 5
5 plo) KAL OLKO
SOLTIOLS EKEL
Q@ucLtaaTNpLo
KW TW Aw COU
OvotacTnpLo 10
ALOQWY OV Ta
PEVEYKNS €
T AUTOLS o.Sapo
6 Bots odAo[KA]y
. s . A ian) ~ .
ens a aa ; =
ees oe OO ee ee ee eee ee
[po]us otko[8]o ES il Ake 11
[ pnlous To [@]uou [y]wy [outro] om 12
[aoT |nptov KU [olovt[at evAloyiv
[Tou] 8u cou Kat Tov A[aov e]t at
[avovois] e[t aut lo yap. [tu ws] av tla]
[okokauTW] 20 pedO[nTe Tov [Lop]
[Hata kw Tl Sav [np ovupe jwv
7 [Ow cou kat 6]u Kat Alevet Kat |
> The ends of lines 21 ff. of col. I are contained in P. Giessen 22. Col. II, 15 ff. is covered
by P. Giessen 19, 34 recto.
I1
LS
I 19-20
P2123
123
[25
II 19
YW 22
Deut 27:21
22
24
PAP. GIESSEN 463
a[is TeXeLoT |y eto[usa
ta k[at day |n ext K[at]
Kat [eudpav |
enon [eyumpo 25
ofev [ku to]u
6v oo[u kat yoa]
wus e[Te|
Our reconstruction.
Cie Ui:
GIL.-R.: [a]Ju(20)[Tov]
The reconstruction of these lines, partially based upon P.G.
22 (to which we had no access), is very doubtful. The above-
mentioned deviation from Gl.-R. in I, 19-20 entailed an
additional change in our reconstruction of lines 20-21, for
which GL.-R. have (20) [tou odokau](21)[Twyata kw TJw.
[TeXeLoTt |nTa. For this reconstruction cf. section VI below.
[eudpav |6non. Our reconstruction.
[we] av: cf. LXX 27:3, 4.
GL-R.: kat Afever].
Pap. Giessen 19, 1.2 verso left
[o Aalos ye[vorTo]
[E]mxatapato[s |
[o] KOLLLWLLEVOS
[Leta adedpns
autou 8vyatplos | 5
TATPOS AUTOU
1 SuyaTpos py
[tp]os [a]utou kat [e]pe[t]
[tas] 0 A\aos YyEVvoLTo
[E |twLKaTapaTos 10
[o] KoLLWLEvaos
Weta TEVvOE
pas AUTOU KAL
[ep leper tras o daols]
[ye ]vot[o] 15
EtiKatap[atos |
o tT[u|tTw[v Tov |
[7A |oLov [autou]
i 20
464 CHAPTER THIRTY
[ |...
[ Kal €plet tas
lo Aaos ye |VoLTO
26 [EtiKatlapat[os] 25
[ | oon
Pap. Giessen 26 recto
Deut 28:20 ekAvoLV Kal TH 1 tovts yn[s cov] Deut 28:2
2 [o]vLoptlov]
Pap. Giessen 26 verso
Deut 28:29 ...vdndadn 1 oovopwo.v kat Deut 28:32
...[T]ugAlog].. 2
P.G. 26 recto
IT 1 [ve |TOv.
Pap. Giessen 13 recto
Deut 29:22 (23) [teotpa|d[ co] 1
[Sopa] Kat [youop]
[pa a8] aula ka oe]
[Bwe |p [as ka]Te
[otpe bev k[s] ev op 5
[yn alutou Kat ev
[6up|w autTou
23 (24) [kat € |povoLv trav
[Ta T]a €6v7 Sta Te
[eto ]LNoEv Ks ov 10
[Tas] ™M yn TavTn
[TL(s) to]xupa 7 opyn
In He]yadn auth
24 (25) [kat e_]mate etre[t]
[Sy klateXettra[v] 15
2 GL-R.: [ws to]yupa = SP fare,
PAP. GIESSEN 465
2. The nature of text contained in P. Giessen
In their detailed description, Gl.-R. suggested that P.G. reflects the
text of Day even though no Zapapeitikév reading identical with any of
the readings of P.G. has been preserved. Gl.-R. supported their view by
referring to the ‘sectarian’ reading ap(?)yap[t]¢tp in 27:4 and to the
agreement between P.G. and SP!4"8 in their renderings of 07171 in 25:7, 8
and of 71n in 29:23(24),
Before analyzing GIl.-R.’s arguments, the text of P.G. is compared
with that of the LXX, the main Greek translation of MT. The analysis
focuses on significant (dis)agreements, distinguishing between more
significant (idiosyncratic) agreements (a) and those less significant (b).
24:17 1 -tpoondutTov. Within Greek literature tpootAuTos, the
main equivalent of 71 in the LXX, is a LXX word.
24:19 “ny - [8]pay[pa]: the main LXX equivalent.
24:21 %oivn - [etavaltp[vynois]. Within the LXX the verb occurs
elsewhere only in Lev 19:10. Since the LXX more frequently
used the equivalents kahapdo@a. and émgvddAtcerv, the
agreement between P.G. and the LXX is even more striking.
27 Inna” - H yuvyn Tou adeAgou avutou, to be contrasted with such
renderings of 02° and nna° as otvvupdos (LXX) and etvyap-
Bpeutns (Aquila).
27:5 ff. ‘4 -Kuptos, based on the Qeré perpetuum of the tetra-
grammaton. See also section V.
27:6 nin>w or3aNx - ALBots odo[KA]n[po]is recurring in the LXX of Josh
9:4 (8:31).
27:22, JOX - yevotto (main LXX equivalent).
2329
Agreements in matters of syntax:
25:7. ox) - eav &e. Note the non-stereotyped rendering of the waw.
25:7-8 andy - (kat) avaBnoeTtat and 27 - [kat] Aadn[covarv] (LXX:
Kal Epovolv), reflecting the same understanding of the
rendering of the tenses.
25:8 7x1 Thy) - [kat otals ett. The rendering of the Hebrew
paratactical construction with the hypotaxis is significant.
466 CHAPTER THIRTY
b.
24:17 mon - exkAtvers. The most frequent equivalent of this root in
the LXX is (€x)kAlvw.
24:17 Yann - evexuptaons, the main LXX equivalent.
257 yon - BovAn6n (LXX: BovAntat), the main LXX equivalent.
27-5 nam - Ovotactnpto, one of the two main equivalents (the
other one is Bwyds).
27:21, 23, WX - eTiKaTapatos, the main LXX equivalent.
24, 26
28:29 wwnn -dmAadn[ (LXX: dynAadnoac).
29:24 lary - [k]atedettra[v].
(25)
The above-mentioned agreements between P.G. and the LXX show that
they reflect the same translation tradition. Sometimes P.G. is even is
closer agreement with that tradition than the LXX ad loc.
Some of the significant agreements between P.G. and the LXX were
already recognized by GI.-R., but their assumption that the Greek
translator of the SP sometimes used the OG does not adequately allow
for the facts. Several of the translation options of P.G. are well rooted
in the LXX tradition, while others betray a close relationship to the
LXX ad loc,
3. P.G. and the LXX
The assumption that P.G. forms part of the textual tradition of the LXx
is not obvious since the differences between the two are numerous. As a
consequence, P.G. cannot be considered representative of the main
tradition of the LXX. P.G.’s place as a revision of the OG is determined
by the following analysis:
a. P.G. brought the OG into quantative and qualitative conformity
with MT.
B. P.G. aimed at greater consistency in translation than the OG.
y. P.G. revised the OG in various other ways.
PAP. GIESSEN 467
a. P.G. brought the Old Greek into conformity with MT®
a. Quantitative changes
2417
2417
24:18
24:18
24:19
24:20
24:21
24:21
25:7
200%
2727.
277
PH fale
DrAD
DIi22
2722
27:22
28:32
29:22
(23)
29:22
(23)
on 1) - TpOONAVUTOU Kat Opdavov | TpOGNAUTOU Opdavou.
MT > - kat xnpas (sub + Syh™8, > x) ] >.
p-xn2 - ev yn AvyuTTw (yn sub + Syh; > d) | [ev aryluttw (based
on calculations of space).
IRN - Gol EvTEdAOLAL (EVT. GoL cmox Arm Spec) | [evtedAopat]
oo[t].
~9- eav 6€ (> b6€ AFMNacdfiklmprxyb?2 Boh La) | Ota[v].
MT > - kat pvnobyon oT. oLKETHS Noba Ev yn ALyuTTTW Sta TOUTO
EYW GOL EVTEAAOPLAL TOLELV TO pnLa ToUTO (> bcfmowx Syh™s &t
Eus) | >.
~D - eav Ge (> S€ Flmb9) |] Or[av].
oin-> - Kat Tw opdavw (> Kat d) |] Tw [opdavu].
nl" 2° - Hn yuvn (+ Tov aSeApou auTou ckptx Syh™S [sub ® vid]
Or") | 7 yuvn Tou adeAdov auTou.
ow nx? op? - avactnoal To ovopa Tou adeAdou autou (Tw
adeAdw avutou ovoyua o) | [avjaotnolar Tw alde€rAb[w av]to[v
ovoua |.
ow - > (exe. APMacfhiklmorxyb, Boh Cyrcodd 1/2 Jee.
MT > - kat €ptANaOnon (> k) | >.
“1 - Aevet (pr kat A Eth) ] kat Afevet].
77171" - lovéa (pr et Eth) | [kat] evo[vda] (calculations of space).
inns - adedgns (THs aSerons autov®, adehdns autou Fdkoptx
Arm Boh) |] a8eAdns autou.
TaN Na - tatpos Bz (ek tps autou N @ cgortxy Eth [pr 7 g Eth), +
avtou n Bohl¥, prek AFM rell Arm) ] €vyatplos] tatpo¢s autou.
IN NI- UNnTpes autou Bbnw (pr ek AFMN 6 rell Arm Eth, autou
q Arm) ] 8vyatpos pn[Tplos [a]uTov.
(nv?D)1 - > (kat Geox) | Kat.
JONI - Ev Bupw (ev opyn autou G’, + autou G3? cfikoptx Syh™8) |
Ev op[yn alutov.
inonl - opyn (pr ev pt, ev opyn auTov cox Syh™8, + avtou AFM
aklmaqrvyb 2) | ev [6ujp]w avtou.
6 The data on the left side of the square brackets refer to the MT and LXX and those on
the right side to P.G.
468
CHAPTER THIRTY
b. Qualitative changes (substitutions)
24:16
24:16
Looe
25¢7.
257,
ZOE
27:4
27-5
27:6
27:6
277
24:19
o°12 1°- texvev (Ta8wv Tht 1/2) ] vw, a more precise rendering,
recurring later in the verse in both the LXX and P.G.
wx -ekaoTtos | [alvnp, a stereotyped rendering also used
frequently in kaige-Th.
yon? x? - un BovdAntat | yn Boudry (note the difference in
aspects).
wn - 0 avOputros ] o avnp (more precise as WX 1s Opposed to the
woman). The LXX revisions usually rendered w°x with avip,
leaving GvOpwtros for OX.
Dx - ett | mpos.
xD - ou BedEL | ouK EBovANGy (note the difference in tenses).
ODnx - cot (usty begn-qtv™&8wxz™8 ay Arm Boh Eth La) ] upwy.
om>y - em auto Bmua? (avTov qy, auTou kx, avta 8, auTots ej
Or8", autoug AFMN rell Cyr®4) ] em avtots.
man min>dw a7ax - ALBous oAoKANpoUS (ALOBoLS OAOKANpoLs 46)
oLKOdSouNnoEts | ALBoLs OAO[KA]N[poL]s orko[S]o[unlots.
spade ‘man (nx mn) - BuoLaatpLov KVELW TW Hew Gou (KU ToU BU
n Arm) ] [O]uci[aot]nptov ku[ Tou] Ou cov.
o-n?w - A@uctav cwtnptou (Qvotactyptov A) | [TeXetot]nTa. The
reviser reduced the frequent LXX rendering of two words to a
single one, and furthermore he turned to a different etymology
of the Hebrew word. See further section VI below, on the
relation to the anonymous marginal notes in codex M of the LXX.
159 - evavttov BNel Cyr®4 1/2 (evavt. AFM@rell Cyr°°44 1/2) ]
[eutr|poodev.
oy -ev |] fe].
pn211y1 - 6btaBavtes (StaBaivovtes hy, S:aBavtwv vuwy dpt Arm
Eth) ] [ws] av tla]pnd8[nte]. Note the stereotyped rendering of
the bet and the pronominal suffix (as in the LXX of 27:3,4).
T8728 10D - Tov VETOV TH yn Gou (THs yn N, Ths yns FP abdefiykm
npqtwx Arm Boh Eth J [ve ]tov ms yu[s cou].
mx - BAevovtat Bqv™8Sz™B8ay (BAETIOVTES U, OUK OWOVTAL ej,
esovta, Au™Syz™8, obovta. FGMN® v'*tz'*trell, videbunt La) ]
OPWOLV.
. aimed at a greater consistency than the Old Greek ad loc.
JY XP WpPN - apnons apnrov (+ variants) ] [Ge ptons] tov B[epropov].
Both renderings are used in the LXX, 8ep{{w 1s more frequent.
24:16
24:16
24:16
24:17
257
2507
PAP. GIESSEN 469
1127(1) - €povotv |] AaAn[covo.v]. The equivalent of P.G.
prevails in the LXX, while eltetv more frequently reflects 7x.
Wann) OXI - Ev OupLw kat opyn (+ variants) ] ev op[yn alutou Kat
ev [6uy]w avtov. The equivalents of P.G. are in closer agreement
with the general practice of the LXX.
. revised the Old Greek in various other ways
MAX PV ww XN? OD) - KaL VLOL OUK aTOBAVOUVTAL UTEP TATE PWY
(+ variants) ovSe vio. uTEp TaTEpwy ato[O@lavouvtar. The
translator of P.G. rendered x?...) (after 8? - ouK) with ouSe (see
next item) and he changed the word order, possibly in order to
create a chiastic pattern. |
N2...1- KGL... ouK (ou8e Arm Or8! Spec6°dd Cyr 2/3 Thdt 1/2) ]
ovuoe.
INONI- Ev TH EAUTOU GLapTLa (auapTLa €auTou ox Arm (vid) Ors",
auaptia avtov d Just Thdt 1/2, dla apaptia ag Cyr 1/2 Thdt©o4
1/2) ] Stats a[ujaptiats. As occasionally elsewhere in the
LXX, t8Lo¢ is used here to represent the possessive pronoun.
MON TQ 2INN N?71- Kal OVK EVE XUPAS LLATLOV yTpas (+ variants)
] Kat xnNpas ouK evexupLaons [tuJatiov. As in 24:16, P.G.
probably created here a chiastic structure. In the new word
order, the tpoondutés, dpdaves, and xfjpa, always mentioned
in one series in Deuteronomy, are now juxtaposed.
Mmywr - ETL THY TUANY (> dn); Tod f; + THS TOAEWS O | ELS THY
Supav. While the equivalent 1yw - TUAn occurs frequently in
the LXX, O0Upa is rarely used for this word. Possibly a place
other than a city gate was meant (in the LXX @vpa is not used
for a city gate), and possibly a gate of a law court was meant
(thus TO, TPs-Jon, and Saadya; see also the anonymous emt To
Kptt... in F® ad loc. and in 17:5)
"na" - 0 ad€AOS TOU av&pos pov ] o ek [Ev |GHepou po[v]. The two
renderings view the kinship from different angles. On the
other hand, nna was left unrevised in this verse.
470 CHAPTER THIRTY
25:8 (197-y) 7371 - N -yepovota } [or oulvetor. Elsewhere in the LXX the
somewhat free equivalent yepouota is used often. Equally free
is the equivalent of P.G., incorporating an element of exegesis,
parallelled by TPsJon and SPT@'8,7 The fact that this render-
ing occurs in SP!4°g is insufficient reason for labeling it
‘Samaritan,’ as was done by Gl.-R., 45—46 and earlier by Field,
Hexapl., I, xxxiii and 330. This exegetical tradition is also
found in Jewish sources.®
29:23, 4X 7IN - 0 Bupos THs opyns | [La]xupa n opyn. The exegetical
(24) tradition that "an means ‘strong’ is shared by P.G., TPs-Jon, TO,
Saadya, SP!4'g, and the Arabic translation of SP. As in the
preceding example, there is no justification for labeling this
translation tradition ‘Samaritan,’ as was done by GL.-R., 53.
In the following two examples, P.G. used equivalents also found in
the later books of the LXX.
24:18 7Ay -otketns | [S]ovAos. The Pentateuch usually used the
equivalent tats, and less frequently olkétns and Sepdtwy,
while the later books of the LXX used mainly So0dA0s and traits.
24:18 727 - pnya |} Ao[yov]. The most frequent equivalent of 127 in the
Pentateuch is pyjua, and Adyos in the other books.
P.G.’s revisions of the underlying LXX were subdivided into three
groups. Attention was first directed to those revisional renderings
which brought the LXX into conformity with MT. P.G.’s quantitative
and qualitative revisions, often coinciding with part of the manuscript
tradition of the LXX, brought the OG closer to MT, while in other cases
(section 3) the reviser allowed himself to be less bound by the Hebrew
text (see section 7 below). The number of revisional renderings is rather
large for the small section covered by P.G.
Since important agreements between P.G. and the OG indicate that
P.G. is part of the LXX tradition, and since P.G. brought the OG into
conformity with MT, P.G. should be considered a revision of the OG. It
is noteworthy that this revision shares with kaige-Th the equivalent
wer - avn (cf. Barthélemy, Devanciers, 47).
7 The fragment which was classified in Brooke-McLean as A4 (Gen 37:3-4,9) reflects a
similar equivalent, viz., copwv, shared with T° and SP!@'8 ad loc. The contention of Rahlfs,
“Ein weiteres Fragment der griechischen Ubersetzung,” NAWG, Phil.-hist. Kl. 1911, 263—
266 that this fragment contains an additional fragment of Zapapettikdv cannot be sup-
ported (for a detailed analysis, see pp. 382-383 of the original article).
8 b. Qidd. 32b Anon APY ~D NOX pt PR. See further J. Levy, Wérterbuch iiber die Talmudim
und Midraschim | (Berlin/Wien 1924) 548. See also the parallelism in Jer 18:18 and Ezek 7:26.
PAP. GIESSEN 471
In the examples listed in paragraphs a and 8, P.G. revised the OG
towards MT when that version is identical with SP. As a consequence,
P.G. could have revised the OG towards SP rather than MT. In three
instances however P.G. revised towards MT when it differed from SP
(see further n. 13):
27:22 WX) - Kal Epovatv (eper Nbejoquv™S8wz™8a. Arm Eth) |] kat
le ]per. Contrast SP 17x).
27:22 MX) - Kal €povoty (epet bfw La) ] kat [e]pet. Contrast SP 17x).
27:23 WX} - Kal Epovotv (eper Nbc“ fiquv™8wz™8a. Arm Eth La
Spec) ] kat [eplet. Contrast SP 1x2.
4. Additional differences between P.G. and the LXxX
Beyond the revisional tendencies described above, there are additional
differences between the two Greek traditions which can be
accommodated within the aforementioned analysis.
a. Synonymous renderings of P.G. and the Old Greek
Synonymous renderings? of P.G. could have replaced earlier renderings
of the OG, as elsewhere in the other revisions of the OG.19
25:7 Tye - emt (THY TuANV) | Ets (THY Bupav).
25:7 MON) - Kal Epet | Kat Ettm (the tense of epet conforms with
that of the preceding avaBroetat; that of etm agrees with
the following ett).
27:4 TWIT... NW - Kkoviacets ... KovLa ] xpta[ts] ... xpLoTNpLW.
27 12 oD72V7 - 6aBavtes (+ variants) | [ws] av m[a]peXO[nTe].
28:20 mown - €KALLLAaV (+ variants) ] ekAvOLV.
29:24 (25) “we 2y - ott (+ variants) ] etre [Ldn].
29:24 (25) atv -katedttTooav B (katTedettTov AF*M®@ (-ev @* vid)
chinoruvyajb9, katedtov F°? G rell ] k]atedetta[v] cf. GL-
R., 54.
? Two renderings are named synonymous when they are found concurrently in the
LXX, not in different grammatical conditions. Many of these synonymous renderings
appear elsewhere as variants in the manuscripts of the LXX. Generally the Greek words
themselves are also synonymous. The term is coined after Talmon’s use of the term
‘synonymous readings.’ For a brief description, see Tov, TCHB, 260-261.
10 Reider, Prolegomena, 34 ff. (Aquila); Hatch, Essays, 27 (Symmachus); Soisalon-
Soininen, Zusatze, 155; Johnson, Rezenston, 42 ff.; J. Ziegler, Bib 40 (1959) 224-227.
472 CHAPTER THIRTY
b. P.G. reflects Hebrew variants
Occasionally the Hebrew Vorlage of P.G. differed from MT. In four of
the following examples, the Vorlage of P.G. differs from both MT and
SP, while in 27:4 it agrees with SP and La against MT. In all five
instances P.G. differs from the LXX.
24:16 MT = SP ixona = LXX ev Tn €avuTov auaptia - EkAtuLav (+
variants) | Stats alpjaptiars = IXDNA.
24:19) MT =SP 47H2 71x; LXX auntov ev Tw aypw oou (+ variants) =
VP TWA ] Tov B[eptopov] Tou aypou cou = Iw Wx.
25:7. MT = SP andy = LXX kat avaBnoetat (> Kat mo Arm Boh) =
xp JIwa )] avaBnoetat = ANY.
27:4 MT %ary 772 = LXX ev oper TarBad (Garzin La!?4) ] ev ap(?)
yap[uJftp!! = SP ovtea3703. While the importance of the
agreement of P.G. with the most important sectarian reading
of SP should not be underestimated, it could also be an ancient
not yet sectarian reading. The fact that the Vetus Latina,
never suspected as Samaritan, preserves the same variant,
points in the same direction, since this source has preserved
many important ancient variants.
29:25, MT = SP = LXX xat epovo tv cf. SP x1 probably to be
(24) vocalized as 19x) (cf. SP!48 pra) | [kau ev]qate = TN).
5. A Samaritan text?
The preceding analysis has attempted to establish that P.G. reflects a
revision of the OG. This assumption is supported by anonymous notes in
manuscript M™6 of the LXX (see below).
While GI.-R. had to allow for some Septuagintal influence on P.G.
(pp. 56-57), they had no doubt that P.G. contains the Greek translation
of SP (Zap). However, the main arguments in favor of this view were
contested in the preceding analysis: ap(?)yap[t](tp in 27:4 (= La and SP)
probably reflects an ancient, not yet sectarian, variant, and the
renderings of O°3?1 in 25:7, 8 and of "7n in 29:23 (24) do not necessarily
1] tt cannot be determined whether apyap([t]{tu was written as one word, as in the
Samaritan tradition. While the transliteration of 17 in 27:4, 12, not paralleled in the LXX,
might point to a Samaritan background (see GI.-R. 48), AppayeSuv (= 1171p 17?) in Rev 16:16
shows the wider use of this transliteration as do many additional translit-erations of
peor apaical terms in the LXX.
12 See especially R.S. Haupert, The Relation of Codex Vaticanus and the Luciantc Text in the
Books of the Kings from the Viewpoint of the Old Latin and the Ethiopic Versions, unpubl. diss.,
Philadelphia 1930. See further the studies mentioned in Tov, “Lucian,”* n. 32.
PAP. GIESSEN 473
reflect Samaritan traditions. The following arguments must be taken
into account as well: (1) P.G. agrees twice with MT and the LXX against
SP.13 (2) P.G. revised the OG three times to MT rather than the SP (see
above). (3) The LXX rendition of the tetragrammaton with kupto¢ is
based on the Qeré perpetuum of the Hebrew in Jewish tradition (7172).
Whether the employment of kuptos in P.G. is based upon the OG or
shows an independent knowledge of the Jewish tradition cannot be
determined (alternatives are S€ométys or 1 AQ as in 4QOLXXLev?). But
the translator of P.G. did not derive this rendering from a Samaritan
source, Since the Samaritans pronounced the tetragrammaton as Shema
(‘the Name’) or lafe.!4 (4) The 43 (46) known readings of Dau agree to
such an extent with SP!4"8, that the former must have derived from the
latter, as Field, Hexapl., I.lxxxiii ff., 329-330 and Kohn!° have shown
in detail. P.G., on the other hand, could not have originated from SP!?'8
(see also Gl].-R., 56, 62).16 Pummer, “Samareitikon” accepts this view as
well. (5) The aforementioned positive evidence connecting P.G. with
the LXxX.
It is hard to give a clear-cut answer to the question whether GI.-R.’s
suggestion should be abandoned. The solution may depend on one’s
understanding of the reading ap(?)yap|[t ty in 27:4 upon which the most
important of GI.-R.’s arguments is based. Neither of the following two
explanations of this crucial reading is completely satisfactory:
1. P.G. is a revision of the LXX, and not a Samaritan document.
ap(?)yap[t]€tu in 27:4 is an ancient variant also found in La, an
important source for such variants. !”
13 94:16 MT max 2 = LXX tatepwv = P.G. ] SP mara.
29:24 (23) MT in = LXX tie = P.G. (considerations of space)] SP 71.
4 Thus Theodoret, Quest. XV in Exod. See Z. Ben-Hayyim, “On the Pronunciation of the
Tetragrammaton by the Samaritans,” Erlsr 3 (1954) 147-154 (Hebr.); W.W. Graf Baudissin,
Kyrios als Gottesname tm Judentum (Giessen 1929), [].217 ff.; J.F. Moore, Judaism (Cambridge
oe 1.426427; J. Macdonald, The Theology of the Samaritans (London 1964) 95, n. 1.
13.5. Kohn, “Samareitikon und Septuaginta,” MGW] 38 (1893-1894) 1-7, 49-67.
16 The following renderings preclude the hypothesis that P.G. was derived from SP!#'8:
24:17 ANYVIIN wi? Jawon x? - Kat xnNpas ovuK eve xupLaons [tuJatLov.
24:19 7732 71NN - Tov 6f[epraj.ov] tov ay [pou cou].
250 Pon) - avaBnoeTtar
25:8 mov 79n") - [kat] Aan [Govowv ] auTw.
25:5 WW") - Ett.
17 For a similar view, see R. Pummer, “APFAPIZIN: A Criterion for Samaritan
Provenance?” JS] 18 (1987) 18-25. Pummer demonstrated that the writing of these two
words as one word occurs also in non-Samaritan sources. The ‘Samaritan’ reading,
involving the continuous writing of the two words, occurs also in a Masada fragment
written in the paleo-Hebrew script (papMas1o). See S. Talmon, “Fragments of Scrolls from
Masada,” Erlsr 20 (1989) 286-287 (Heb. with Eng. summ.). The Samaritan nature of that
474 CHAPTER THIRTY
2. P.G. represents a Samaritan revision of the LXX carried out for
internal and/or external needs of the Samaritan community. A possible
parallel to such a practice 1s provided by Abu Sa‘id’s thirteenth
century Arabic translation of SP, based on the earlier translation of MT
by Saadya.!8 A further parallel is provided by the close relationship
between SP!4'8 to TO which has been interpreted variously.!?
The latter view is more problematical than the former one: (1) By
implication the Samaritan reviser took the OG as his base, adapting
and revising it to SP; (2) The first two of our arguments against GI.-R.
would still be valid. However, one might contend that the Vorlage of
the reviser differed from the text of SP known to us; (3) The argument
stressing the close connection between Zap and SP!4'8 versus the
noticeable lack of such connection between P.G. and SP!4'8 remains
valid. Otherwise one would have to postulate the existence of two
different Greek translations of SP; (4) The agreement between SP, P.G.
and La in the crucial reading o°1772 remains problematical.
The first view of P.G., preferred here as the simpler one, may be
strengthened by the possible connection between P.G. with an
anonymous revision of the LXX to be mentioned in section 6.
6. The relation between P.G. and individual manuscripts of the LXX
Because P.G. is considered to be part of the LXX tradition, agreements
between individual LXX manuscripts and P.G. were analyzed in our
original article (1971, pp. 377-380). While not close to any of the known
witnesses of the LXX of Deuteronomy, P.G. shares revisional readings
with Arm and Syh™ (both representing the Hexapla), with manu-
scripts bdnw (possibly Lucianic manuscripts),2° and with manuscripts
Fceox, containing a proto-Hexaplaric recension.?!
fragment is contested by H. Eshel, “The Prayer of Joseph, a Papyrus from Masada and the
Samaritan Temple on API'APIZIN,” Zion 56 (1991) 125-136 (Heb. with Eng. summ.).
18 See E. Robertson, “The Relationship of the Arabic Translation of the Samaritan
Pentateuch to that of Saadya,” in: E. Rosenthal (ed.), Saadya Studies (Manchester 1943)
166-176; Kahle, Catro Geniza, 53-55.
196. Kohn, Samaritanische Studien (Breslau 1868) 4 ff. and L. Goldberg, Das Samaritanische
Pentateuchtargum (Bonn 1935) 30 ff. assume that SP!#"8 was interpolated from T°. Other
opinions are referred to by Kohn, ibid., and Waltke, Prolegomena, 75 ff.
20 Thus D.W. Gooding, Recensions of the Septuagint Pentateuch (London 1955) 15: “In Deut.
Lucian’s recension is to be found, if anywhere, in® gn bw and possibly dpt.’ J.W. Wevers,
Text History of the Greek Deuteronomy (MSU XIII; Gottingen 1978) 17-30 uses the neutral
term ‘the n group.’
Thus Gooding, Recensions, 9 ff. (see n. 20 above).
PAP. GIESSEN 475
While the known revisions of the LXX ad loc. reveal no special
relationship to P.G., a few important points of contact with an
anonymous revision of Leviticus should not remain unnoticed. The main
parallel is between Deut 27:7 o-n?w - [tedetot]nta (LXX: Ovatav owtn-
ptov) in P.G. and anonymous marginal notes in M™'S (Paris Bibl. Nat.
Coislin 1 of the 7th century CE) in Lev 3:3, 9; 4:10, 26 mentioning
TEAELOTNS next to cwTTpLov of the LXX. There seems to be no better
reconstruction of P.G., and in view of the fact that this equivalent does
not recur elsewhere,** the link with M™4'8 is remarkable. The fact that
M™arg contains many revisional renderings and often agrees with
equivalents of P.G. (for an analysis, see pp. 380-381 in the original
article) further strengthens the link with this revision of the OG. The
link between M™'8 and SP!2'8 was also pointed out by Wevers.”?
7. Character of the revision
The anonymous reviser attempted to reflect his Hebrew Vorlage as
precisely as possible, but not to the same extent as kaige-Th or Aquila.
On the one hand he used Hebraistic renderings such as dv7p for wx, but
on the other hand he allowed himself such freedoms as the translation
of 71 with ouvetéos.
The only clue for determining the date of the revision embodied in
P.G. is provided by the date of the fragments which were dated by Gl.-
R. to the fifth-sixth century CE. The revision itself was earlier.
22 Cf. similar equivalents in Judg 20:26B, 21:4B (redela) and 2 Chr 29:35 (Tedelwats).
23 J.W. Wevers, Leviticus, Septuaginta, Vetus Testamentum graecum auctoritate academiae
scientiarum gottingensis editum (Gottingen 1986) 31. See further Pummer, “Samareitikon,”
397-404 (see n. 2 above).
CHAPTER THIRTY-ONE
LUCIAN AND PROTO-LUCIAN
TOWARD A NEW SOLUTION OF THE PROBLEM
Much new material on the Septuagint revisions has been revealed
through archaeological finds and philological investigations. The
present paper deals with one of these recently discovered revisions
—the so-called proto-Lucianic revision. Since any analysis of this
version is by its very nature closely related to Lucian’s revision, we
shall first outline the three major opinions proffered on the nature of
his revision and the essence of the manuscripts boce9.
From Ruth 4:11 onwards, Lucian’s revision is contained in bocye), as
was suggested by Rahlfs on the basis of earlier suggestions by de
Lagarde. Our remarks are limited to 1-4 Reigns since they have been
the subject of the majority of investigations concerning Lucian’s revision.
A. Rahlfs’ thorough study Lucian’s Rezension! formed the basis of
the communis opinio on Lucian until two decades ago. Rahlfs described
how, on the one hand, Lucian brought the OG into conformity with the
Hebrew, while, on the other hand, he removed the OG from MT by
freely revising its language and style. Rahlfs further realized, as had
earlier scholars like Mez,* that Lucian’s fourth century revision reflects
many ancient variants, which Rahlfs named proto-Lucianic since they
are also to be found in various sources preceding Lucian by several
centuries. It was Rahlfs’ great achievement to have described the three
layers composing Lucian’s text. As a rule, he underestimated the
importance of proto-Lucianic elements.°
1 See the analysis by N. Fernandez Marcos, “The Lucianic Text in the Books of
Kingdoms: From Lagarde to the Textual Pluralism,” in: Pietersma—Cox, De Septuaginta, 161-
174.
2 A. Mez, Die Bibel des Josephus untersucht fur Buch V-VIII der Archaeologie (Basel 1895).
3 Cf. PLL. Hedley, “The Gittingen Investigation and Edition of the LXX,” HThR 26
(1953) 69: ’Rahlfs has always admitted that Lucian may have used a Syrian text that
differed from those current in other districts, but he has consistently depreciated the value
of the recension.’
478 CHAPTER THIRTY-ONE
A completely novel view of the nature of bocze2 was suggested in 1963 by
Barthélemy, Devanciers, 89 ff. After describing the characteristics of
the newly discovered kaige-Th revision, Barthélemy turned to an
analysis of the second part of 2 Samuel in the Greek versions. He
showed that in this section the main LXX manuscripts contain kaige-
Th, while the OG is found in bocze. In order to prove this hypothesis,
Barthélemy showed that bocze2 and the other manuscripts have a
common basis, and he further demonstrated that kaige-Th revised the
tradition embodied in bocze2 in conformity to the Hebrew. Indeed,
kaige-Th is more literal than bocze2, but this situation does not
necessarily imply that kaige-Th revised bocje2. The relationship
between kaige-Th and bocze2 could be viewed differently:
1. Barthélemy’s examples are selective and exclude those showing
that bocze2 are more literal than kaige-Th.
2. There is much internal evidence in bocze2 indicating that they
contain a revision, even in the second part of 2 Samuel.4
3. Barthélemy’s conclusions refer to the whole of the LXX, while his
investigation is limited to one section of Reigns.
4. Barthélemy dismisses the historical evidence concerning Lucian’s
revisional activities with too much ease.°
A third view of bocge2 was proposed in 1964 by Cross, “Biblical
Text.” While analyzing 4QSam4, Cross realized that this Hebrew
source contains many proto-Lucianic readings.® In light of this evidence,
Cross suggested that bocze2 are composed of two different layers: a
substratum containing a proto-Lucianic revision of the OG toward a
Hebrew text such as 4QSam4, and a second layer containing the
historical Lucian’s corrections. These ideas were reinforced by Lemke,”
Shenkel, Chronology, Klein,® Harrington,’ and O’Connell, Exodus.
While agreeing with the position that bocyge2 are composed of two
layers,!° I would question whether the substratum is indeed a proto-
4 Cf. especially S.P. Brock, “Lucian redivivus, Some Reflections on Barthélemy’s Les
Devanciers d’ Aquila,” in F.L. Cross (ed.), Studia Evangelica V (TU 103) 176-181 as well as his
study quoted in n. 33.
> Cf. Cross, “Biblical Text,” 295, n. 44.
Cross has published only some examples of proto-Lucianic readings of 4QSam; see
further DJD XVII (in press).
7 W.E. Lemke, “The Synoptic Problem in the Chronicler’s History,” HThR 58 (1965) 349-
363.
8RW. Klein, “New Evidence for an Old Recension of Reigns,” HThR 61 (1968) 492-495.
7 D.J. Harrington, “The Biblical Text of Pseudo-Philo’s Liber Antiquitatum Biblicarum,”
CBQ 33 (1971) 1-17.
The fact that diametrically opposed tendencies are visible in bocge2 makes such an
assumption very plausible.
LUCIAN AND PROTO-LUCIAN 479
Lucianic revision. Has it really been established that this substratum
was a revision rather than simply another Greek text? If such an
assumption is necessary to explain the elements in bocze, which
approximate the LXX to MT, it must be pointed out that Lucian derived
such elements mainly from the ‘Three’ and the fifth column of the
Hexapla, as shown by Rahlfs.
In light of the above reflections and of my own study of Lucian, a new
working hypothesis on the nature of bocze2 is suggested here. Like
Cross, I propose that boce2 in the books of Reigns are composed of two
layers. The second layer is the historical Lucian, and I suggest that its
substratum contained either the OG translation or any OG translation.
Although the term proto-Lucianic (or pre-Lucianic) should probably be
continued as a designation of the OG substratum, one should not assume
a proto-Lucianic revision as such, since the existence of such an
intermediary stage has not been proven. Our proposal thus forms a
compromise between the views of Barthélemy and Cross.
The working hypothesis is supported by the following arguments:
1. The text of bocze2 is evidenced in a wide range of sources, both
before and after Lucian’s supposed floruit (300 CE). Of these sources, the
sources preceding Lucian are particularly noteworthy. These include:
4QSam2 (see Cross, “Biblical Text’),
two early papyri: the Manchester P. Ryl. Greek 458 of Deuterono-
my?! and P. 2054 of Psalms,
the various fragments of the Vetus Latina,!
the substratum of the Armenian translation,!%
the text quoted by Josephus,!4
the text quoted by Pseudo-Philo, Biblical Antiquities,!°
11 C£. the literature quoted by Kahle, Cairo Geniza, 220-223 and in addition: J. Hempel,
ZAW NF 14 (1937) 115-127; A. Allgeier, Biblica 19 (1938) 1-18; J. Hofbauer, ZKT 62 (1938)
385-389.
123M. Metzger, Chapters in the History of New Testament Textual Criticism (Leiden 1963)
31-32 mentions several studies on the relationship between Lucian and the Vetus Latina.
See further: Wellhausen, Samuel, 221-224; H. Voogd, A Critical and Comparative Study of the
Old Latin Texts of the First Book of Samuel, unpubl. diss. Princeton 1947; R. Thornhill, JTHSt 10
(1959) 233-246; J. Cantera, Sefarad 23 (1963) 252-264; idem, “Puntos de contacto de la
‘Vetus Latina’ con la recension de Luciano y con otras recensiones griegas,” Sefarad 25
ee 69-72; cf. also Brock’s study mentioned in n. 33.
3 Cf. B. Johnson, Die armenische Bibeliibersetzung als hexaplarischer Zeuge im 1. Samuelbuch
(Lund 1968) 158; idem, “Some Remarks on the Daughter Versions of the Septuagint,”
ee 5:(1972) 7-9.
4 CE A. Mez, Die Bibel von Josephus (Basel 1895); A. Rahlfs, Lucian’s Rezension, 80 ff.;
H.St.J. Thackeray, Josephus, The Man and the Historian (1929; repr. N. Y. 1967) 87 ff.
1S cf, D.J. Harrington, S.J., “The Biblical Text of Pseudo-Philo’s Liber Antiquitatum
Biblicarum,” CBQ 33 (1971) 1-17.
480 CHAPTER THIRTY-ONE
the text quoted by various Church Fathers: Clemens of Alexandria,!®
Theophilus of Antioch (cf. Rahlfs, Lucian’s Rezension, 114 ff.), Tertul-
lian,!7 Hippolytus (see Rahlfs, Lucian’s Rezension, 123 ff.), Cyprian,!®
and Origen.!? To these sources one should probably add the Coptic
translation of the LXX?° and certain elements in the Peshitta.*!
Contrary to the beliefs of some scholars,** I disagree with the
opinion that the enumerated sources have been retouched by so-called
‘Lucianic revisers.’ In some instances such an assumption is either
impossible or close to impossible.
The list of sources which reflect the text of bocye» after the
historical Lucian is equally large. It contains both the text quoted by
various Church Fathers? and the text reflected in the Gothic, Slavo-
nic, and so-called Syro-Lucianic translations of the LXX. Some of these
sources are undoubtedly based on Lucian, but in other cases it has yet to
be determined whether the post-Lucianic sources are based on the
ancient substratum of bocze2 only or whether they reflect the Lucianic
text as a whole. This is especially true since some of these sources are
very close to Lucian’s floruit and/or are not derived from the area of
Antioch.
It cannot be coincidental that so many diverse sources reflect a proto-
Lucianic text in the books of Reigns. The only solution appears to be
that all the above-mentioned sources reflect elements of either the OG
or a single OG translation underlying Lucian’s revision. The non-
Lucianic manuscripts contain a different, and, sometimes later, text
tradition. We shall later dwell on the differences between bocze7 and
the non-Lucianic manuscripts.
2. The studies dealing with the character of the assumed proto-
Lucianic revision stress that this revision generally left the OG
16 cf. Barthélemy, Devanciers, 136 (pace Rahifs, Lucian’s Rezension, 118 ff.).
17 cE. Pp, Capelle, Le texte du Psautier latin en Afrique = Collectanea biblica latina 1\V (Rome
ae 200 (pace Rahlfs, Lucian’s Rezension, 138 ff.).
18 cf. especially B. Fischer, “Lukian-Lesarten in der Vetus Latina der vier Koénings-
bucher,” Miscellanea biblica et orientalia R.P. Athanasio Miller oblata = Studia Anselmiana
XXVII-XXVIL (Rome 1951) 169-177; Capelle, Le texte, 203-204.
19 Rahlfs, Lucian’s Rezenston, 139 ff.; D. Barthélemy, Devanciers, 136 ff.
20 cf. J.B. Payne, “The Sahidic Coptic Text of J Samuel,” JBL 72 (1953) 51-62; however, it
is not certain whether the Old Coptic text is as early as Payne surmises (250 CE).
21 See the data collected by Th. Stockmayer, ZAW 12 (1892) 218-223; however,
Stockmayer’s conclusion, which is phrased in the title of his article (“Hat Lucian zur seiner
Sep ue pma revision die Peschito benuitzt?”: yes!), cannot be vindicated.
2 See especially L. Dieu, “Retouches lucianiques sur quelques textes de la vieille version
latine (J et IJ Samuel),” RB NS 16 (1919) 372-403.
Asterius Sophista, Diodore of Tarse, Eustathius, Lucifer, Ambrose, Augustine,
Theodore of Mopsuestia, John Chrysostom, Theodoret, Jacob of Edessa.
LUCIAN AND PROTO-LUCIAN 481
unrevised (see especially Shenkel, Chronology). This view is based
upon a comparison of translation options in bocze7 and the OG which
shows that both have a common vocabulary, on Shenkel’s investigation
of the identical chronological systems of the two as opposed to the
chronological systems of MT and kaige-Th (ibid., passim), and on the
fact that both traditions start the third book of Reigns at 1 Kgs 2:11.74
But if the assumed proto-Lucianic revision is so close to the OG and
frequently left it unrevised, would it not be more logical to characterize
the substratum of bocze2 as OG rather than a proto-Lucianic revision?
3. It has been recognized (e. g., Barthélemy, Devanciers, 128 ff.) that
the contents of the Hexapla’s sixth column in the second part of 2
Samuel are very close to bocze2. For Barthélemy the sixth column thus
contains the OG, while the other manuscripts in that section contain
kaige-Th. For Cross, “Biblical Text,” 295 it contains the proto-Lucianic
revision ‘in relatively pure form.’ Since in this section Origen placed
kaige-Th in the fifth column, it would be more likely that the
Hexapla’s sixth column would contain the OG than an unknown revision
about which we possess no records. Furthermore, a probable parallel is
found in 2 Kings, where Burkitt suggested that the Quinta contains the
OG. As in the second part of 2 Samuel, the main manuscripts of 2 Kings
contain kaige-Th, and here, also, bocze2 resemble one of the columns of
the Hexapla, in this case the Quinta. However, the resemblance
between bocze on the one hand, and the sixth column on the other, has
yet to be investigated in detail.
4. Klein2® has shown that the additions in the Greek Chronicles
harmonizing that text with 2 Kings reflect the textual tradition of
bocze, rather than kaige-Th.2’ This situation would seem to indicate
that the translator of Chronicles took the OG as his basis.?8
5. Shenkel demonstrated that in the synoptic sections of Samuel and
Chronicles the Greek Chronicles is based on the OG of Samuel.*? He
found that in the kaige-Th sections there is a much greater agreement
24 Cf. Shenkel, Chronology, 10 ff. and section 2 of the appendix to this paper.
> F.C. Burkitt, “The So-called Quinta of 4 Kings,” Proceedings of the Soctety of Biblical
Archaeology 24 (1902) 216-219.
6 Klein, “New Evidence” (n. 8).
The alternative explanation that the historical Lucian harmonized the two Greek
texts has been discussed by Klein and Allen in HThR 61 (1968) 483-495.
8 This possibility, which seems to us the best explanation of the evidence, is rejected by
Klein after some consideration: ‘This pre-kaige text, which served as the source for the Par
supplements, could be either the Old Greek hitherto unknown, or, as seems more likely, the
proto-Lucianic recension’ (ibid., 104).
23) 1D. Shenkel, “A Comparative Study of the Synoptic Parallels in I Paraletpomena and
J-II Reigns,” HThR 62 (1969) 63-85.
482 CHAPTER THIRTY-ONE
with bocze2 of Samuel than in the non-kaige sections.29 An analysis of
Shenkel’s data proves that in all sections the Greek Chronicles is based
upon the OG, which in the kaige-Th sections is reflected in boc ep.
On the basis of these arguments suggesting that the substratum of
bocze2 contains the OG or OG elements it seems that the study of bocze>
should be founded on a new basis. The contents of bocye» should be
studied anew, especially in light of the proto-Lucianic sources. While
the above-mentioned arguments partially relied on previous research,
a renewed investigation of boczez may support the working hypothesis
suggested above.
When starting to elucidate the details of the working hypothesis in
light of the above-mentioned arguments, one realizes the difficulties in
defining criteria for unraveling the three layers of bocze9, viz. the OG
substratum, Lucian’s borrowings from the ‘Three’ and the fifth column of
the Hexapla, and Lucian’s own corrections. Criteria have to be defined
as to which elements belonged or could have belonged to any one of the
three layers. One of the main problems is that certain characteristics of
bocy2e2 which scholars have always assigned to the historical Lucian
were already extant in Lucian’s Vorlage. B. Fischer showed in 1951 (cf.
n. 18) not only that the so-called Lucianic tendencies were already
extant in La, but also that this version—no doubt an early translation of
the OG—reflected some of these tendencies against bocze2 and the other
manuscripts of the LXX. Thus, additions of subjects, objects and names,
changes between nouns and pronouns, short contextual additions,
harmonizing additions, specific translation equivalents, several
doublets, some linguistic changes, and translations instead of
transliterations*! are not late Lucianic phenomena, but belonged to the
very first stratum of the LXX. To these examples one may add several
in which bocze2 reflect the original Greek text which has been
30 According to Shenke]’s statistics, 74% of the words of 1 Chronicles 17-18 agree with
the Old Greek of 2 Samuel 7-8 (non-kaige) and 3% disagree with the Old Greek in favor of
boc2e2 in that section. On the other hand, 56% of the words in 1 Chronicles 19 agree with
the kaige manuscripts of 2 Samuel 10 and 13% disagree with the same manuscripts in favor
of bocze9. The amount of agreement between 1 Chronicles 19 and the bocze2 manuscripts
in 2 Samuel 10 is actually much greater than the numbers indicate: 1. the 56% agreement
between 1 Chronicles 19 and the kaige-Th manuscripts of 2 Samuel 10 include many
instances when kaige is identical to bocze9; it would actually be more correct to state that
in such cases 1 Chronicles agrees with bocge of 2 Samuel 10 while the kaige revision has
left the text unrevised. 2. The number of assumed agreements between 1 Chronicles 19
and the OG of 2 Samue! 10 would have been larger than 13% if the historica] Lucian had
not inserted his own revisions in boc7e9.
3] Even though it appears illogical for original renderings to have been changed by a
later hand to transliterations (mainly of unknown words), I hope to have established this
process for some of the LXX revisers in Tov,”Transliterations.”*
LUCIAN AND PROTO-LUCIAN 483
corrupted in all other manuscripts. The most striking examples of proto-
Lucianic elements are provided by readings which reflect early
variants. These variants, which often are of importance for Biblical
scholarship, form the largest group of proto-Lucianic readings (see the
appendix).
Since it can be proven that certain readings of boc2e2 were extant in
pre-Lucianic sources, we have to consider the possibility that many
typologically similar readings were pre-Lucianic as well. This point
should not be overstressed, however, since it is not impossible that in
some cases the historical Lucian was guided by the same principles as
the original translators.
In concluding this point, the importance of La and other pre-Lucianic
sources in pinpointing the ancient elements of bocge2 should be
emphasized.%4 An investigation into the first stratum of bocye> is of
primary importance for solving the enigma of bocze9.
A second line of investigation attempts to pinpoint readings which
the historical Lucian derived from the ‘Three’ and from the Hexapla’s
fifth column. This investigation is limited by its very nature since
Lucian’s sources have been preserved only partially, but probably the
majority of the quantitative revisions towards MT are derived from the
‘Three’ of the fifth column. The second category of proto-Lucianic
readings in the appendix shows that some of Lucian’s quantitative
revisions may have been found in his Greek Vorlage.
The changes which Lucian himself introduced have to be further
examined. One immediately thinks of several Atticizing changes such
as the replacement of Hellenistic forms as eftav and éAdfooav with
eltov and €daBov, Ajpwer with Ajwer, and of the replacement of the
passive aorist éyev6n with the middle aorist éyéveto.%4
Lucian probably introduced certain stylistic corrections, such as the
insertion of synonymous words. However, much investigation remains to
be carried out in this field as well; a comparative study of translation
equivalents in the different sections and manuscripts of the LXX should
32 See J. Cantera, “La Vetus Latina y e] Texto Masoretico, Hipotesis de una revision de la
Vetus Latina a base del texto hebreo,” Sefarad 23 (1963) 252-264; J.C. Trebolle Barrera, “Old
Latin, Old Greek and Old Hebrew in the Books of Kings,” Textus 11 (1984) 17-36; idem,
“Old Latin, Old Greek and Old Hebrew in the Books of Kings (1 Ki. 18:27 and 2 Ki. 20:11),”
Textus 13 (1986) 85-95; N. Fernandez Marcos, Scribes and Translators—Septuagint and Old
Latin in the Books of Kings (VTSup 54; Leiden/New York/K6ln 1994); idem, “The Vetus
Latina of 1-2 Kings and the Hebrew,” in: Greenspoon—Munnich, VIII Congress, 153-170.
33 SP. Brock, The Recensions of the Septuagint Version of I Samuel (Oxford 1966; repr. Torino
1996) discusses several Atticizing and stylistic changes by the historical Luctan. See further
N. Fernandez Marcos, “Literary and Editorial Features of the Antiochian Text in Kings,” in:
Cox, VI Congress, 287-304.
484 CHAPTER THIRTY-ONE
determine which bocze2 synonyms were introduced by Lucian and which
were included in his Vorlage. In this respect also, La and other pre-
Lucianic sources are of help.*4 All these studies can now be based on new
editions of the Lucianic (Antiochene) text in 14 Reigns.*°
In conclusion, it is suggested here that the substratum of boc ye,
contains either the OG translation or any single OG translation. The
non-Lucianic manuscripts contain kaige-Th in two sections in the books
of Reigns, and in three sections they reflect a text which is usually
described as the OG. This suggestion is acceptable for the kaige-Th
sections: the OG is contained in the substratum of bocze2, while kaige-
Th and the second stratum of bocye) reflect later corrections of this old
substratum. However, how should one explain the relationship
between bocze2 and the other manuscripts in the non-kaige sections? It
appears that in these sections we should continue to characterize all
non-Lucianic manuscripts as the OG. But in the instances in which
bocze2 deviate from the other manuscripts, a different view of the
latter manuscripts is suggested for which two alternatives should be
considered:
1. In his preliminary publication of 4QLXXLev4,37 Skehan has
shown that in many details this text reflects the OG, while all extant
manuscripts have been retouched. This situation shows the
unreliability of the manuscripts of the LXX, especially when they
reflect MT exactly. The situation in the historical books parallels the
problem raised by 4QLXXLev?: whenever the bocze2 reading is at
variance with MIT or renders it freely, the reading found in the
remainder of the manuscripts, as a rule, agrees with MT, and could thus
represent a later revision. In other words, in the non-kaige sections the
substratum of bocze2 always represents the OG, while the other
manuscripts as a rule reflect the OG, but at times their text may have
been retouched.
34 For example, of the few stylistic changes which Brock assigned to Lucian in his article
mentioned in n. 4, two are already evidenced in La (cf. the translations of o1%9H in 2 Sam
11:7 and of *n in 2 Sam 11:11).
° N. Ferndndez Marcos and J.R. Busto Saiz, El texto antioqueno de la Biblia griega, I-II
(Madrid 1989, 1992); B.A. Taylor, The Lucianic Manuscripts of 1 Reigns, Volume 1, Majority
Text, Volume 2, Analysis (HSM 50,51; Atlanta, GA 1992, 1993).
Since bocze2 and the other manuscripts of the LXX are genetically interrelated in all
sections of the books of Reigns, it is likely that the kaige-Th revision was based upon the OG
substratum of boc2e7. However, it has yet to be determined whether this substratum of
boc2e2 was identical to the assumed bsloe! S of kaige-Th or whether one should posit two
closely related OG traditions (OG! and OG?)
7PW. Skehan, VTSup 4 (1957) 148-160. Similarly the final publication (DJD IX).
LUCIAN AND PROTO-LUCIAN 485
2. It may be suggested that both the bocje2 reading and the one found in
the other manuscripts represent two parallel OG traditions.
Since the relationship between the readings of boce2 and that of the
other manuscripts may, as a rule, be described as that between an
original and its revision, one may prefer the first possibility. However,
at the present stage of knowledge of the proto-Septuagint question it is
hard to solve this problem. The substratum of bocze2 is therefore
characterized as containing either the OG or any OG translation. The
latter possibility allows for the existence of other translations that
might be grouped with the OG while different in some details.
As suggested above, the existence of a proto-Lucianic revision of the
LXX has not been established. It is further suggested that the
substratum of bocyze2 contains either the OG or any single OG
translation. The correctness of the first suggestion does not hold an
implication for the validity of the second.
APPENDIX
SOME CATEGORIES OF PROTO-LUCIANIC READINGS REFLECTED IN
MANUSCRIPTS BOC2E? IN REIGNS
1. Proto-Lucianic variants (* MT)—the majority of the proto-Lucianic
readings fit into this category.
Zeal 13 MT agi
4QSam? jp")
boe, Syz) (txt) ’ Lwvvabav
B*ao *lwvadsap
Ba(vid)b AMN rell Arm Co Eth Thdt ’TwvadaB
Jos. Ant. VII, 178 "Twvainc
cf. Cross, “Biblical Text,” 294
2 Sam 18:2 MT TW nw
boz™8c,e9 Kal €tploceuce AavetS (717 ww)
Lxxrell Kal dtTeéaTerAev Aaverd
Lay et tripartitum fecit
486 CHAPTER THIRTY-ONE
2. bocze2 = La = MT #LXX —the historical Lucian could have derived
these readings from the ‘Three’ or the Hexapla’s fifth column, but since
they are reflected in La, they may have been original.
1 Sam 1:6 MT nyt Naya
bogheg Chr(vid) 81a 10 €E€oubevetv alti
LXxrell —
La” quia ad nihilum reputabat eam
2 Kgs 17:21 MT yy?
borc7e> éEppayn
Lxxrell =
La©yPr dissipatus est
3. Ancient doublets
1 Sam 16:14 MT Innylz)
bocze, Syn) ouvetxev... Kal ETVLyEV avTéV
Lxxrell ETTVLYEV avTdv
La comprehendit ... et soffocabat eum
Cf. further B. Fischer, “Lukian-Lesarten ...”, 177.
4. bocye2 reflect the OG from which the corrupt text form of the other
manuscripts has developed
1 Sam 9:24 MT OUT 7K?
Nabe-osvwyz™8bc2e2!*t Syh) ™8 Thdt Tapa tot Aaob
(ayn nxn)
B Tapa ToUG ddAouG
A Tapa Tov aVou
La® =bocye, — (acc. to Brooke-McLean)
1 Sam 10:2 MT 71909
bioc»e2 év (tote) dploic
BV To Oply
LXxrell EV TU OPEL
Laby in finibus
1 Kgs 18:32 MT Tun
Oc*»e Qaaha
bz™8c4,? Saraa
dip Oadaav
Lxxrell 6dAaccav
LUCIAN AND PROTO-LUCIAN 487
5. boczé2 contain a translation while the other manuscripts contain a
transliteration
2 Sam 17:29 MT V2 mpwr()
a™Sbgoz™8c7e7 Arm yaka@nva Locydpta
Lxxrell cadwé Bodv (or sim.)
Lab vitulos saginatos
La¥ lactantes vitulos
Clem. Alex. I, 98 Woo xdp.a yakabrnvd
Cf. further B. Fischer, “Lukian-Lesarten ...”, 176.
6. boczeé add subjects or objects (these additions could reflect variants)
J Sam 10723 MT —
bhoxbyc72e7 Co ZA\LOUNA
Lxxrell __
Labv Samuel
1 Kgs 21:20 MT —
(20:20) bioc2e2 *HALaG
LXxrell _
LaLuc Helias
Cf. further B. Fischer, “Lukian-Lesarten ...”, 176.
7. Contextual additions (these additions could reflect variants)
1 Sam 30:15 MT —
Mbgioy®zarc7e> Arm kal Wpogev auto
LXxrell —_
La¥ et turavit et David
1 Sam.9:3 MT —
bdghiopc ze Kal dvéoTtn 2aovad Kal TapéAa-
Bev ev Tév trarsaplwv Tov
TATPOG aVTOU WET avTov kal
eTrope vO Cytetv Tac Svouc Kis
ToU TaTpoG avuToU
LxXxrell __
S yo IN? Toy 797) TN) PNW OP
“TIINT NVR RVING NM
488
CHAPTER THIRTY-ONE
8. The internal division of the books of Reigns
The following sources of III Reigns start at 1 Kgs 2:12: bocze2, MS Vat.
Syr. 162 (cf. A. Rahlfs, Septuaginta-Studien Il, 16, n. 1), Josephus, Ant.
Jud. (book VII includes 1 Kgs 1:1—2:11), Diodorus, Theodoret, Syv). Cf.
further A. Rahlfs, ibid., 186 ff.
9. Translation technique
1 Sam 9:27
2 Sam 11:7
1 Sam 4:4
MT 13/(2)
b’b™tozMEcre, Arm dkpov
Lxxrell [LE POG
Lab in loco summo
MT ovr ows) axr awd (717 2xw7)
Mann ow’)
boc.e2 el bytatvet ‘IwaB kal el vyratvet
adc Kal el Vytalvet 6 ThE LOG
LxXxrell elc elphvnav “Twa kal ele elpfhvnv
tov Aaov kal ele elpyvny Tot Toe LOU
La recte est Joab et recte est populus et
recte est exercitus belli
MT eles bee ta ee
boz??cse> ov éTekdOnTo Ta XepouBiy
Lxxrell Kafnuéevou XepouBety
Labv ubi sedebat in cherubin
10. Grammatical changes
1 Sam 1:19
1 Sam 2:25
MT
bocr.e
LXxrell
La™
MT
WIAD... VANw
TpooeKUVTOaP ... Kal €TOpEeVBNoav
Kal Tpookuvotoat ... Kal TOPE VOVTAL
adoraverunt ... abierunt
ON}
bozcze2 Sy) Or Chr éav 8
LxXxrell
LavCypr
Kal €av
si autem
CHAPTER THIRTY-TWO
THE METHODOLOGY OF TEXTUAL CRITICISM IN JEWISH GREEK
SCRIPTURES, WITH SPECIAL ATTENTION TO THE PROBLEMS IN
SAMUEL-KINGS—THE STATE OF THE QUESTION!
Samuel-Kings have always drawn the special attention of scholars
interested in textual problems. In few books does the ‘LXX’ reflect so
many interesting Hebrew readings, very often superior to MT, as in 1-2
Samuel, and in addition 1 Reigns offer the student of the LXX a
wealth of inner-Greek problems.
Interest in 1-4 Reigns increased when the find of 8HevXlIlgr drew
Barthélemy’s attention to the different hands discovered by
Thackeray in 1+ Reigns, and when Hebrew manuscripts of Samuel from
Qumran provided Cross with new insights into textual problems of both
the Hebrew and Greek Bible.
The publication of Barthélemy, Dévanciers (1963) and Cross,
“Biblical Text” (1964) inaugurated a new period of understanding of the
problems of the Greek 1-4 Reigns, and of the whole Greek Bible. Both
studies have provided us with a wealth of new data as well as with
stimulating new theories.
Beyond the problems discussed with regard to all the books of the
LXX, the four books of Reigns have raised several major issues, mainly
in the field of recensional activity.
1. The possible distinction of different translation units in 14 Reigns
and a description of their character.
2. The relationship between the kaige-Th sections in Reigns and
other witnesses of the kaige-Th group.
3. The characteristic features of the kaige-Th revision.
4. The relationship between Aquila, Symmachus, and kaige-Th.
5. The problem of bocze2, the Old Greek, Lucian, and proto-Lucian.
! This paper was presented at a symposium at the SBL congress in Los Angeles in 1971
on “The Methodology of Textual Criticism in Jewish Greek Scriptures, with Special
Attention to the Problems in Samuel-Kings.”
490 CHAPTER THIRTY-TWO
6. The relationship between the Greek and Hebrew texts in 14
Reigns.
7. The synoptic problem of the Greek texts of 1-4 Reigns and 1-2
Paralipomena.
8. The nature of the sixth column of the Hexapla in 14 Reigns.
9. Text and midrash in the third book of Reigns.
1. The unity of 1-4 Reigns
After Thackeray had discovered that different translators were invol-
ved in the translation of the prophetical books, his attention was
drawn to 1-4 Reigns where he uncovered a similar pattern. In both
cases Thackeray pointed to manifold Hebrew words, roots and
expressions which are represented differently in two or three sections of
the same book, indicating, according to Thackeray, that different
translators had rendered these sections. In 1-4 Reigns, Thackeray
distinguished five such sections, of which the third (By) and the fifth
(y5) were rendered by one translator. Contemporary scholarship is still
much indebted to Thackeray’s pioneering studies, at first formulated in
the separate articles mentioned in notes 2-3, and later in a monograph.*
Thackeray’s examples of differences in translation equivalents
between the various sections are, as a rule, correct. His description is
also valuable as it contains many insights, such as his conclusion that
the translator of By and y& was a faithful translator, close to
Theodotion—Barthélemy went one step further, suggesting that the
two are, in fact, identical.
In his study of 1-4 Reigns, Thackeray noticed important differences
in translation technique between the individual sections of 14 Reigns.
He did not pay attention to important agreements between the different
sections, i.e. translation options which are characteristic of the four
books of Reigns.? One should be able to uncover such agreements between
the OG sections of Reigns and the OG substratum of the kaige-Th
sections when the text has remained untouched by the kaige-Th
reviser. Since Thackeray did not define any idiosyncratic agreements
between the individual sections of 14 Reigns, he did not suggest that
2 “The Greek Translators of Jeremiah,” JTS 4 (1902/3) 245~266; “The Greek Translators of
Ezekiel,” ibid., 398-411; The Greek Translators of the Prophetical Books,” ibid., 578-585.
See also “The Bisection of Books in Primitive Septuaginta Manuscripts,” JTS 9 (1907-8) 88~
98.
3 “The Greek Translators of the Four Books of Kings,” JTS 8 (1906-1907) 262-278.
4 The Septuagint and Jewish Worship (Schweich Lectures 1920; London 1923).
3 In Jeremiah, on the other hand, Thackeray noticed some agreements between the two
sections of that book, cf. JTS 4 (1902/1903) 253-254.
METHODOLOGY OF TEXTUAL CRITICISM 49]
the relationship between the different sections is one of translations
and revision. Barthélemy, however, did suggest such a view, and it
therefore appears that his study ought to have investigated the
problem of the agreements in order to establish a relationship of
‘source’ - ‘revision.’
By approaching the problem from a completely different angle,
Barthélemy did provide a partial answer to the above-mentioned
problem. By suggesting that kaige-Th revised the tradition embodied
in boc,e, and that the latter tradition is the OG, Barthélemy
established the ‘source’ - ‘revision’ relationship in a different manner.
The same apples to my own suggestion (Tov, “Lucian”*) that boc,e2 do
not incorporate the OG in a pure form, but that their substratum
contains the OG upon which Lucian’s revisions were superimposed.
Most scholars agree that 14 Reigns do not form a unity and that the
extant manuscripts are composed of juxtaposed sections of the OG and
kaige-Th. In view of the mentioned complications it is not impossible
that Thackeray’s views are still supported by some scholars (see T.
Muraoka, “The Greek Texts of Samuel-Kings: Incomplete Translations
or Recensional Activity?” presented together with the present paper
[see n. 1]).
The attention of scholars has been directed to the question of why
our manuscripts display a mixed text, at times original and at times
revised. Thus, why does section By start in the middle of a book (at 2
Sam 11:2 according to Thackeray and Barthélemy) and end at 1 Kgs
2:11, and why does section yé start at 1 Kings 22 and not at 2 Kings 1?
As for the incipit of section By, Thackeray, “Kings,” 263, asserted
that it had been purposely omitted by the first translator because of its
contents (‘the story of David’s sin and the subsequent disasters of his
reign’) and filled in by a later translator. Barthélemy, Dévanciers, 141,
similarly suggested that kaige-Th retouched only this section because
of his interest in its contents. Shenkel, Chronology, on the other hand,
started section By at 2 Sam 10:1, and not at 11:2, on the basis of textual
evidence rather than the contents of the chapters, admitting that ‘the
reason for beginning the KR at 10:1 is not yet apparent’ (ibid., 118).
Similarly, no plausible solution has been suggested for the incipit of
section yS. Thackeray again ascribed the change of text type to
theological factors, while Barthélemy, Dévanciers, 42, followed by
Shenkel, Chronology, 63, suggested that section y6 started at 1 Kings 22
because the LXX inverted the order of the preceding two chapters.
Since previous theories could not explain adequately the alternation
of unrevised and revised sections in the manuscripts of the LXX, it may
492 CHAPTER THIRTY-TWO
be suggested that the phenomenon originated for purely mechanical
reasons. The OG translation of Jewish Scriptures required several
scrolls, and large books like Samuel—Kings were usually not written in a
single scroll (see the Qumran evidence). Supposedly the archetype of
the extant manuscripts of 14 Reigns was composed of scrolls consisting
of different text types,° probably because the compiler of this
archetype was unable to obtain scrolls of the same type, or was unaware
of their mixture. Various parallels and arguments may be adduced in
support of this suggestion (see Tov, Jeremiah and Baruch, 162).
2. The relationship between the kaige-Th sections in Reigns and other
witnesses of kaige-Th
Barthélemy, Dévanciers, 47, lists the various members of the kaige-Th
group. The list is very impressive indeed, including, among other
things, the various attestations of Theodotion, 8HevXIIgr, the third
and fifth sections of the “LXX’ of Reigns, and the so-called Quinta
column of the Hexapla. The group should probably be called kaige-Th
rather than simply kaige because in antiquity Theodotion’s name was
attached to this anonymous revision.
Monographs have been devoted to the different members of this
group. Barthélemy, Dévanciers, 89 ff., offered a start in providing a
detailed study of the By sections of Reigns. Subsequently Shenkel,
Chronology described the Greek text of 1-2 Kings and O’Connell,
Exodus. investigated Theodotion’s revision of Exodus. Among other
things, Shenkel supported Barthélemy’s suggestion that the By and yd
sections are part of the kaige-Th revision. A similar conclusion was
reached by O’Connell with regard to the Theodotionic version of
Exodus chosen because of the relatively large number of attested
readings of Theodotion. W. Bodine, The Greek Text of Judges—
Recensional Devel-opments (HSM 23; Chico, CA, 1980) established
that the B text of Judges is a member of the kaige-Th group. These
monographs added a wealth of new data and views. Shenkel, for
example, described in extenso the chronological system of MT reflected
in Greek in kaige-Th, as opposed to the Greek system of chronology,
embodied in the OG translation and in the bocze2 text in the By and y&
sections.
6 2 Reigns was contained in two different scrolls (2 Sam 1:1—10:1; 2 Sam 10:2—1 Kgs
2:11). The Greek 4 Reigns may have started at 1 Kings 22, but it is not impossible that the
division between 3 Reigns and 4 Reigns was determined by the length of the scrolls. For a
similar suggestion, see R.A. Kraft, Gnomon 37 (1965) 282-283.
METHODOLOGY OF TEXTUAL CRITICISM 493
The various attestations of kaige-Th are presently dispersed, but
Barthélemy believes that all witnesses display one and the same
revision.’ It is hard to determine whether the group indeed forms a
unity, as Barthélemy claims.
Schmitt, Theodotion (1966) provides data for undermining the
hypothesis that the kaige-Th group forms a homogeneous unit. Since
Schmitt has convincingly demonstrated that the sixth column of the
Hexapla derived from a different source from the so-called Theodotion
translation of Daniel, the kaige-Th group must have been composed of
at least two layers.
3. The characteristic features of kaige-Th
As a means of determining the identity of supposed members of the
kaige-Th group, Barthélemy described several translation equivalents
characterizing this group. The most characteristic translation
equivalent of this group is that of O01 withkatye, explained by
Barthélemy in accordance with the hermeneutical rule of ribbuy amit,
‘inclusion and exclusion,’ which refers to certain Hebrew particles that
are always presumed to include at least one element in addition to the
word(s) mentioned after it. All other characteristic translations of
kaige-Th are also explained by Barthélemy in the light of rabbinical
sayings or hermeneutical traditions, mainly from the Mekhilta, e.g.
the translation of wx - ‘everyone’ with dvyp, 7218 with éya elut, and the
etymological translation of the roots 2%1/2x°. Barthélemy probably
went too far in his desire to explain all renditions of kaige-Th in
accordance with rabbinic exegesis. It may very well be that all the
renditions he discussed—with the possible exception of 0) - kalye—
simply represent a very literal, root-linked translation technique in
which each Hebrew root is represented by its fixed translation
equivalent. While the translational approach may have been
influenced by certain hermeneutical rules of the Rabbis—cf. rather the
school of Agiba than the school of Ishmael frequently quoted by
Barthélemy as an illustration for kaige-Th—an attempt to explain all
the characteristic renditions of kaige-Th in accordance with rabbinical
sayings seems far-fetched. Thus, no quotation from rabbinical sources is
needed to explain the stereotyped rendition of w°x with dup including
those cases in which the Hebrew is used in the sense of ‘everyone.’
7 Barthélemy, Dévanciers, 47 mentions, however, a few witnesses to a later revision of the
LXX which is closely related to kaige-Th.
494 CHAPTER THIRTY-TWO
After Barthélemy’s list, additional translation equivalents charac-
terizing kaige-Th have been described by Shenkel, Chronology, 13 ff.,
113 ff.; Smith,? Grindel,? O’Connell, Exodus, 286-291, and Tov, “Trans-
criptions,”* and “Lucian.”* Earlier studies of Driver, Samuel, iv ff., de
Boer, Schmid,!9 Kelly,!! Gehman,!* and Wevers! provide general
descriptions of the translation technique of the different sections of 14
Reigns.
4. The relationship between Aquila, Symmachus, and kaige-Th
In two different sections of Dévanciers, Barthélemy refers to the
relationship between Aquila, Symmachus, and kaige-Th. First, he
provided examples of Aquila’s dependence upon kaige-Th in the choice
of certain translation equivalents (Dévanciers, 81-88). Secondly,
together with his description of 8HevXIIgr, Barthélemy describes the
relationship between this scroll (belonging to the kaige-Th group),
Aquila, and Symmachus (pp. 246 ff). He arrives at the conclusion that
Aquila and Symmachus did not revise the OG translation of the Bible,
as expected, but revised kaige-Th. Kaige-Th was thus not simply a
forerunner of Aquila, but provided the very text upon which Aquila and
Symmachus made their improvements. Barthélemy’s conclusion thus
explains a feature which has not been dealt with previously: why do
the margins of several LXX manuscripts contain so many readings
common to the ‘Three’ or two of them? The agreements cannot have been
coincidental, and some kind of close relationship must have existed
between the ‘Three.’ We now realize that kaige-Th was the source for
both Aquila and Symmachus, and when this source was left unchanged
by these translators, the reading could have been annotated in the
margin of a manuscript as ‘the Three.’
8M. Smith, “Another Criterion for the Kaige Recension,” Bib 48 (1967) 443-445.
9 J.A. Grindel, “Another Characteristic of the kaige Recension: nsh - nikos,” CBQ 31 (1969)
499-513.
10 p AH. de Boer, Research into the Text of I Samuel I-XXI (Amsterdam 1938); idem, “1
Samuel XVI. Notes on the Text and the Ancient Versions,” OTS 1 (1942) 79-103; idem,
“Research into the Text of I Samuel XVIII-XXXI,” OTS 6 (1949) 1-100; J. Schmid,
Septuaginta-geschichtlische Studien z. 1. Samuelbuch, unpubl. diss. Kath.-Theol. Fakult. (Breslau
1941).
MBH. Kelly, The Septuagint Translators of I Samuel and IT Samuel J:1-/:1, unpubl. diss.
Princeton Theological Seminary (Princeton, N.J 1948).
12 HS. Gehman, “Exegetical Methods Employed by the Greek Translator of I Samuel,”
JAOS 70 (1950) 292-296.
J). Wevers, “Principles of Interpretation Guiding the Fourth Translator of the Book of
the Kingdoms,” CBQ 14 (1952) 40-56; id., “A Study in the Exegetical Principles Underlying
the Greek Text of 2 Sam 11:2—1 Kings 2:11,” CBQ 15 (1953) 30-45.
METHODOLOGY OF TEXTUAL CRITICISM 495
At the time, Barthélemy’s conclusions were not completely convincing
since they were based on isolated readings of the ‘Three’ which could
not be judged within the context of whole sentences and sections.
However, complete sentences and sections of Theodotion do form the
basis of O’Connell, Exodus which has shown convincingly that Aquila’s
revision of that book was based on Th-Exodus. The same can be
demonstrated for 1-4 Reigns since the running text of Aquila, published
by Burkitt,!4 can now be compared with kaige-Th in section y8.!> The
relationship between the ‘Three’ can further be studied in the
Hexaplaric Psalms fragments published by Mercati.!©
5. Manuscripts bocze, the OG, Lucian, and proto-Lucian
There is no consensus about the nature of bocze in Reigns. This question
is significant for the understanding of 1 Reigns, its manuscripts, and in
fact, for the whole LXX. While the problem of the nature of bocze? is
not limited to the four books of Reigns, it is most obvious in these books.
The present comments are limited to the nature of bocze2 in the four
books of Reigns, because the evidence found in these books differs from
that in the other books. This limitation is further justified for practical
reasons because the greater part of previous pilot studies on boczey and
the Lucianic problem have been carried out on 1-4 Reigns. All
conclusions reached should be applied to these four books only, with
subsequent investigation determining how far these conclusions are
applicable to other books.
Barthélemy, Dévanciers, not only offered new insights on kaige-Th,
but also dealt in length with the relationship between bocze and the
other manuscripts in the By section of Reigns. The second issue was
reopened by Barthélemy. Tov, “Lucian’”* describes four positions on this
issue: A. Rahlfs, Lucian’s Rezension (1911); Barthélemy, Dévanciers
(1963); Cross, “Biblical Text” (1964); Tov, “Lucian”* (1972).
The nature of the tradition(s) embodied in bocze, need to be further
examined, especially with the aid of concordances of those
manuscripts:
1. The nature and quantity of pre-Lucianic elements in bocye.
14 EC. Burkitt, Fragments of the Books of Kings according to the Translation of Aqutla
(Cambridge 1897).
et studies performed on the Burkitt text support Barthélemy’s assumptions.
6G. Mercati, Psalteri: Hexapli Religuiae (Vatican 1958); cf. also J. Venetz, Die Quinta des
Psalteriums. Ein Beitrag zur Septuaginta- und Hexaplaforschung (Massorah, Série 1, vol. 2;
Hildesheim 1974).
496 CHAPTER THIRTY-TWO
2. A description of the many sources reflecting pre-Lucianic elements.
Each source provides a number of specific problems, e.g. La and 4QSam?.
3. Criteria for isolating the revisional layer of bocze2 and a descrip-
tion of its nature.
4. The nature of bocye, in the kaige sections (By and y&) as compared
with its nature in the non-kaige sections.
6. The relationship between the Greek and Hebrew texts in 1-4 Reigns
Traditionally, 1-2 Samuel is referred to as the Septuagintal source of
greatest value for the textual criticism of the Hebrew Bible as it
contains a relatively high number of variant readings which are
superior to MT. It also contains valuable material for the literary
criticism of that book, such as the LXX’s shorter account of the story of
David and Goliath (see Tov, “Samuel”*). Much important data are
also found in the LXX of 1-2 Kings, especially in manuscripts bocze2
whose chronological system is preferred by some scholars (see Shenkel,
Chronology; Miller?”).
The Qumran manuscripts support some of the emendations previously
suggested. While the studies of Wellhausen, Samuel, Thenius,1® and
Driver, Samuel, have not been superseded, the overall picture of the
value of the LXX for the textual criticism of these books will be
reconsidered in light of the Qumran manuscripts (see Tov, “Qumran’’*).
For the question of which text is reflected by 4QSam4, 4QSam>, and
4QSam‘, all of which show some form of proximity to the LXX—see
Tov, “Qumran.”*
7. The synoptic problem of the Greek texts of 1-4 Reigns and 1-2 Parali-
pomena
The synoptic problem of the Greek texts in the historical books refers to
the synoptic texts in the so-called ‘LXX’ and in bocze7, with regard to
their relationship to the Greek and Hebrew texts ad loc. and in the
parallel passage.
Differences between the Greek translations in the parallel sections
cause nO surprise since we are faced with different translations of
different Hebrew Vorlagen. Noteworthy are special agreements
7 J.M. Miller, “The Elisha Cycle and the Accounts of the Omride Wars,” JBL 85 (1966)
441-454; “Another Look at the Chronology of the Early Divided Monarchy,” JBL 86 (1967)
276-288.
18 ©. Thenius, Die Biicher Samuels erklért (Leipzig 1898).
METHODOLOGY OF TEXTUAL CRITICISM 497
between the two Greek translations, both in vocabulary and in the
details of the Hebrew text (against MT). It has therefore been
suggested that one of the two translators used the parallel translation,
or that one of the two translations has been secondarily harmonized.
The former opinion is supported by Rehm!? who provided much
plausible evidence that the Greek translator of Chronicles used the OG
of Samuel in the non-kaige sections and the bocze2 text in the kaige
sections. This shows (see Tov, “Lucian”*) that the Greek translator of
Chronicles always used the OG of Samuel, which, in the kaige sections,
is reflected in the substratum of boc2e2. The latter opinion is reflected in
studies by Gerleman?9 and Allen.*! The problem of the relationship
between the Greek texts of 1-4 Reigns and 1-2 Chronicles needs further
investigation.
A special problem is raised by the harmonistic additions to the
Greek text of Chronicles, adding elements from the parallel passages in
2 Kings. Klein has rightly shown that these additions follow the
textual tradition of bocye. in 2 Kings rather than kaige-Th.** He
concluded: ‘This pre-kaige text, which served as the source for the
Paralipomena supplements, could be either the OG hitherto unknown,
or as seems more likely, the proto-Lucianic recension.“ Despite Allen’s
counter-arguments (n. 21) attempting to demonstrate an inner-Lucianic
harmonization in the parallel texts, it seems that Klein is right in
assuming that the translator of Chronicles used the tradition embodied
in bocze2 in the kaige section in 2 Kings, which reflects the OG rather
than a proto-Lucianic revision.
In addition to the specific problems mentioned above, the overall
problem of the relationship between the synoptic texts in bocge2 needs
to be reinvestigated beyond the study by Rehm (n. 19) which demon-
strated the close relationship between the ‘Lucianic’ parallel texts.
19M. Rehm, Textkritische Untersuchungen zu den Parallelstellen der Samuel-Konigsbucher und
der Chronik (ATAbh 13:3; Miinster i. W. 1937) 34ff.
20 G. Gerleman, Studies in the Septuagint I, Chronicles (Lund 1946) 37 ff.; Synoptic Studies in
the Old Testament (Lund 1948) 30 ff.
2117. Allen, “Further Thoughts on an Old Recension of Reigns in Paralipomena,” HThR
61 (1968) 483-491.
v R.W. Klein, Studies in the Greek Text of the Chronicler, unpubl. diss., Harvard University,
Cambridge, MA 1966.
23 RW. Klein, “New Evidence for an Old Recension of Reigns,” HThR 60 (1967) 93-105;
61 (1968) 492-495. The quotation is from HThR 60 (1967) 104.
498 CHAPTER THIRTY-TWO
8. The nature of the sixth column of the Hexapla in 1-4 Reigns
It has been recognized that the Hexapla’s sixth column (‘Th’) in Reigns
By is close to bocze2. For Barthélemy, Dévanciers, 128-136, the sixth
column thus contains the OG, while the other manuscripts in that
section contain the kaige-Th revision. For Cross, “Biblical Text,” 295, it
contains the proto-Lucianic revision ‘in relatively pure form.’ The
nature of the sixth column in Reigns needs to be studied in detail,
especially for sections By (described in general terms by Barthélemy,
Devanciers, 142-143) and y8.
At the same time, renewed attention needs to be directed to the so-
called Quinta of 2 Kings for which Burkitt suggested in 1902 that it
contains the OG.*4
9. Text and midrash in the third book of Reigns
The Greek text of 1 Kings differs greatly from its Hebrew counterpart: it
omits parts, adds elements, contains important duplicate translations
(cf. especially 1 Kings 2), its text is differently arranged and its
chronological system differs from that of MT. Gooding has described
these discrepancies,*> trying to demonstrate that they are not isolated
phenomena, but are part of a deliberate scheme of re-ordering.
Gooding’s summarizing article (“Text and Midrash”) discusses the
question at which level the changes entered the LXX. He concludes
that the majority of the changes were probably inserted by a reviser of
the LXX: “This revision was probably based, at least in part, on written
Hebrew (or Aramaic) traditions of one kind or another’ (p. 2). These
traditions may be pinpointed in some cases in rabbinic sources and
Gooding provisionally calls them ‘haggadic midrash.’ Gooding’s
monograph on 1 Kings (Gooding, Relics) further described the deviating
tradition of the Greek text of 1 Kings.*6
24 F.C. Burkitt, “The So-called Quinta of 4 Kings,” Proceedings of the Society of Biblical
Archaeology 24 (1902) 216-219.
49 “ Ahab according to the Septuagint,” ZAW 35 (1964) 269-280; “Pedantic Time-tabling
in the 3rd Book of Reigns,” VT 15 (1965) 153-166; “The Septuagint’s Version of Solomon’s
Misconduct,” VT 15 (1965) 324-335; “An Impossible Shrine,” VT 15 (1965) 405-420;
“Temple Specifications: A Dispute in Logical Arrangement between the MT and the LXX,”
VT 17 (1967) 143-172; “The Septuagint’s Rival Version of Jeroboam’s Rise to Power,” VT 17
(1967) 173-189; “Text-Sequence and Translation-Revision in 3 Reigns IX 10 - X 33,” VT 19
(1969) 448-469.
26 See also Z. Talshir, The Alternative Story of the Division of the Kingdom—3 Kingdoms 12:24
a-z Jerusalem Biblical Studies 6; Jerusalem 1993).
METHODOLOGY OF TEXTUAL CRITICISM 499
The methodological problems and the new facts about 14 Reigns have
wide implications for study of the LXX as a whole and for the textual
criticism of the Hebrew Bible:
1. The complexity of the relationship between the various Hebrew
textual traditions and the Greek traditions, with special attention
being paid to the existence of non-MT manuscripts underlying Greek
traditions, is emphasized.
2. An analysis of kaige-Th and bocze further clarifies the growth,
composition, and transmission of the LXX.
3. The alternation of different text types in the Greek manuscripts of
1-4 Reigns underscores the impression that the present collection of LXX
books is an amalgam of different text types, late and early, original
and revised.
CHAPTER THIRTY-THREE
TRANSLITERATIONS OF HEBREW WORDS IN THE GREEK
VERSIONS
A FURTHER CHARACTERISTIC OF THE KAIGE-TH REVISION?
1. Background
Theories tend to be perpetuated in research, especially in textual
studies. Such perpetuation may be observed in descriptions of the
translation techniques of Theodotion. Both ancient and modern
presentations of Theodotion stress that Theodotion had much in
common with the LXX! and that he is best characterized by his
tendency to transliterate Hebrew words instead of translating them.
The latter characterization depends chiefly on Field’s analysis,?
which includes a list of 110 transliterations,? and is corroborated by a
remark of Jerome’s on Isa 19:15.4
It was inevitable that the labeling of Theodotion as the
transliterator par excellence would influence scholars in their opinions
about transliterations in the LXX. Consider, for example, the theory of
C. Torrey, who noted a wealth of transliterations in the LXX of
1 Cf. Swete, Introduction, 43.
2 Field, Hexapl., xxxix: ‘Styli autem proprietas, qua Noster a ceteris interpretibus maxime
differt (my italics, E.T.), mos est ejus voces Hebraeas, etiam eas quarum translatio non ita
difficilis erat, dvepynvevtoucg relinquendi, Graecis tantum characteribus pro Hebraicis
positi.” Field’s analysis and terminology were influenced by the description of Theodotion
Py A B. de Montfaucon, Hexaplorum Origents quae ee (Lipsiae et Lubec 1769) 128-130.
3. Torrey, Ezra Studies (Chicago 1910) 69: ’... Tneodotion’s chief characteristic’; Swete,
Introduction, 46 ‘... His chief defect’; H.StJ. Thackeray, The Septuagint and Jewish Worship
(London 1921) 14: ‘... best known for his habit of transliteration’; J. A. Montgomery, A
Critical and Exegetical Commentary of the Book of Daniel (N. Y. 1927) 40-41; Kahle, Cairo
Geniza, 254-255; Schmitt, Theodotion, 56-59; Jellicoe, SMS, 83. Cf. also the various handbooks
to the text of the Hebrew Bible. For the understanding of this habit of copying information,
it is noteworthy that the number of examples in Field’s list (110) is quoted as 90 both by
sera Introduction and by Roberts, Text, 125.
4 Theodotio more suo ipsa verba Hebraica posuit chephpha et agmon (P.L. 24, 254C).
Jerome’s statement on Theodotion has not influenced the research of recent centuries;
since it has been quoted only by Schmitt, Theodotion, 57, n. 1.
502 CHAPTER THIRTY-THREE
Chronicles-Ezra-Nehemiah and therefore concluded that these books
must have been rendered by Theodotion.°
The number of transliterations listed by Field has impressed many
scholars, but it appears that a good many of them should be removed
from the list:
1. Transliterations that are common to the LXX and Theodotion
versions of a certain verse, especially when they occur passim in the
LXX such as pavaa and apafa, both of which are attested several times
for Theodotion as well. In some cases the LXX and Theodotion present
different forms of the same transliteration, indicating that the
transliterations might have been made independently.
2. Transliterations occurring in Th-Daniel. It has been shown
convincingly by Schmitt, Theodotion that Theodotion and Th-Dan
cannot be identified as the same individual. As a result, examples for
Theodotion should not be quoted from Th.-Dan (see also n. 31).
3. Collective readings such asa’@ ora‘o’@ ’(e.g., Isa 13:22 a‘o 6’
LLL). Since both Aquila and Symmachus are known to have transliter-
ated Hebrew words,® Field should not have included these examples in
his list, even though by using his intuition he may have acted
correctly.? Moreover, many collective readings are imprecisely
transmitted.®
4. Anonymous readings denoted ‘anon.’ by Field,? even though some
readings are not marked as such, e.g. Judg 8:26 otpuivwy (cf. Field and
Brooke-McLean ad loc.).
9 Torrey, Ezra Studies, 66-84. While the theory was already outlined before Torrey
(especially by H. Howorth, see Torrey, ibid., 16), he was the first scholar to develop this
argument in detail and to provide a (highly exaggerated and unclassified) list of
transliterations in Chronicles-Ezra-Nehemiah. Jellicoe, SMS, 290-294, describes Torrey’s
thesis tn detail. Torrey’s views on the transliterations in Chronicles-Ezra-Nehemiah are
contradicted and convincingly refuted by B. Walde, Die Esdrasbiicher der Septuaginta,
(Biblische Studien xviii; Freiburg im Breisgau 1913) 37 ff.; G. Gerleman, Studies in the
Septuagint, II: Chronicles (Lund 1946) 1-7; R.W. Klein, Studies in the Greek Text of the
Chronicler, unpubl. diss. Harvard University (Cambridge, MA 1966) 311 ff.
For a more moderate view on this issue, see Thackeray, Grammar, 31. Thackeray assumed
that many transliterations in the LXX of Judges—2 Chronicles and 2 Esdras (i.e., the
transliterations only) ‘are probably derived from him (scil. Theodotion).’
6 Field, Hexapl., xxiii, n. 33; xxxiii n. 16; Reider, Prolegomena, 19-20.
7 Barthélemy, Devanciers, 246-53, 261-265, and O’Connell, Exodus have made it plausible
that Aquila and Symmachus did not revise the Old Greek, but improved upon kaige-Th. If
proved correct, this opinion may pinpoint the origin of many if not most of the collective
readings as kaige-Th.
8 See Reider, Prolegomena, 11-15; P. Katz and J. Ziegler, “Ein Aquila-Index in Vorbe-
reitung,” VT 8 (1958) 264-285; Barthélemy, Devanciers, 27 and passim.
A few anonymous readings may be attributed to Theodotion in accordance with other
readings by him.
TRANSLITERATIONS 503
According to our count, Field’s original number of 110 transliterations
should thus be reduced to 64. This change in numbers is not drastic, nor is
it a necessary link in our argumentation, because not all the
transliterations of Theodotion have been preserved (see n. 15). It
merely puts in relief the narrowness of the evidence on which Field’s
description 1s based.
2. Theodotion
A second criticism of Field’s characterization of Theodotion refers to
the relation between the number of transliterations in Theodotion and
in the LXX. If Theodotion is considered the transliterator par
excellence, the number of transliterations transmitted for Theodotion
should by definition outnumber those of the LXX. However, there is no
basis for such a comparison since the transliterations of the LXX and
their distribution have not been systematically investigated.!°
Attention should be paid to the uneven distribution of the trans-
literations in the books of the LXX. If transliterations of proper nouns
are excluded (see below), some books (the Torah and the Hagiographa
with the exception of the historical books and Canticles) contain only a
very small number of transliterations or none at all. At the other
extreme stand the books Judges-Chronicles, among which 2 Kings
presents the largest number,!! 30 according to our count (see below).
The number of transliterations in 2 Kings is relatively larger than in
Theodotion. Thus, if the translator of 2 Kings had rendered the whole
Bible, transliterating by the same ratio, his transliterations would
have outnumbered by far those of Theodotion, as far as known, even if
all of Field’s 110 examples are upheld. In view of this, Field’s claim
that transliterating is characteristic of Theodotion becomes untenable.
10 For some remarks, see Thackeray, Grammar, 31-38; Swete, Introduction, 324-325; F.
Wutz, “Die Bedeutung der Transkriptionen in der LXX,” BZ 16 (1924) 193-203, esp. 194;
idem, Die Transkriptionen von der LXX bis zum Hieronymus (Stuttgart 1933). Cf. further N.
Sumota’s (incomplete) list of transliterations, AI AMETA®PAZTO! AEZEIZ EN TQ KEIMENQ
TON O’ (Salonika 1969), based on HR.
11 This fact has not remained unobserved, see Thackeray, Grammar, 31; A. Rahlfs,
Septuagintastudien III (Gottingen 1911) 85, n. 2; G. Gerleman, Studies, 7; J.W. Wevers,
“Principles of Interpretation Guiding the Fourth Translator of the Book of the Kingdoms (3
K. 22:1-4 K. 25:30)," CBQ 14 (1952) 42-43.
The relatively large number of transliterations in 2 Kings is not justified by the number of
hapax legomena or rare words in that book, since that book does not contain a larger number
of hapax legomena than the other books. Cf. M. Schloessinger, “Hapax Legomena,” JE 6
(1904) 226 ff.
504 CHAPTER THIRT Y-THREE
It should at least be pointed out that the transliterations of 2 Kings
outnumber those of Theodotion.
In view of the fact that both the LXX of 2 Kings and Theodotion!4
contain a remarkable number of transliterations, one wonders whether
the two Greek versions have something in common, and this question
can now be answered in the positive. Barthélemy, Devanciers, 47, and
Shenkel, Chronology proved beyond doubt that the “LXX’ of 1 Kings 22-
2 Kings (Reigns yd of Thackeray) does not present the Old Greek
translation of Kings, but a subsequent revision which Barthélemy has
called kaige-Th. Transliteration must have been inherent in the
revision technique as understood by this reviser.
3. Classification
A classification of the types of transliterations of the LXX and
Theodotion helps us to determine why this technique was adopted. The
transliterations may be grouped in four categories, each of which has a
different background:
(1) proper nouns;}3
(2) technical terms;!4
(3) words probably unknown to the translator, which thus remained
untranslated. All these are either hapax legomena or very rare (see
Tov, “Loan-words”*);
(4) Transliterations of common nouns erroneously transliterated as
proper nouns because of the context (such as lists of names, cf., e.g., the
transliteration of o7°ny o-377) in 1 Chr 4:22).19
The statistical analysis pertaining to the internal distribution of
these four groups in the LXX and Theodotion is based upon HR on the
one hand, and Field’s list of Theodotion’s transliterations on the other
hand. Except for transliterations of unknown words, we count the
lexicographical incidence of transliterations and not the actual number
of their occurrences. The transliterations of proper nouns (10 in
12 Not all the transliterations of Theodotion have been preserved (the preserved
evidence pertains mainly to Job, Isaiah, and Ezekie]).
3 Personal names, geographic and ethnic names, both single and compound.
4 The majority of these are in the field of religion and architecture or are measures and
weights. Technical terms may occur passim in the LXX (e.g., xepouBe.v, pavaa), or
occasionally (e.g., va¢etp[atoc], Judg 13:5, 7; 16:17).
The preponderance of such transliterations in 1 Chronicles and 2 Esdras is readily
explained by the fact that these books contain several lists of names.
TRANSLITERATIONS 505
Theodotion and a very large number in the LXX) is disregarded, because
by their very nature they are a necessary feature of any translation.
The conclusion of the analysis is that in both the LXX and
Theodotion, transliterations of unknown words form the largest group
(LXX: 108; Theodotion: 46; cf. the appendix), followed by transliter-
ations of technical terms (LXX: 39; Theodotion: 8) and by group 4 (16 for
the LXX).
That the largest group of transliterations should be that of unknown
words!® is compatible with the revisional character of kaige-Th.!7 In
the case of such words, in contradistinction with conjectural renderings
used by the Old Greek translators (see Tov, “Understand”*), the reviser
apparently preferred to retain the original form by leaving the word
untranslated (for examples, see Tov, “Understand”*). In fact, when
comparing Theodotion’s transliterations with the renditions he
replaced, one notes that the great majority of these renditions are
translation guesses.!® The practice of transliterating was considered
adequate by revisers who looked for precision!’, since it left room for
them or for later generations to replace the transliteration29? with a
correct rendition.! It is true that this assumption presupposes a rather
16 Our explanation differs from that of Kahle, Cairo Gentza, 254-255 for the
transliterations of Theodotion: ‘names of animals, plants, garments and all sorts of technical
terms’ which were transliterated because ‘the Greek-speaking Jews were familiar with
such Hebrew words.’ While this reasoning is correct with regard to technical terms, it does
not adequately account for the transliteration of rare, unknown, or difficult words. Kahle
apparently was unaware that the majority of Theodotion’s transliterations represent hapax
legomena or are very rare in the Bible.
The relationship between the two main groups of transliterations in Reigns yé
(unknown words 24: technical terms 6 = 4:1) resembles that between the two groups in the
whole LXX (107:39, that is, approximately 3:1).
8 The transliterations of Reigns y& (kaige) cannot be compared to the Old Greek
translation of that section because it has not been preserved, nor to the parallel accounts in
Chronicles which lack the majority of sections in which the difficult words occur.
The appendix lists for the LXX (a) unknown words which were left untranslated,
and (b) transliterations of unknown words which were probably understood as proper
nouns. For Theodotion no examples of subgroup b could be found. The reason for the
transliteration of words listed in both groups is the same. However, the examples of
a ite b (5 for 2 Kings) are not indicative of precision on the part of the translators.
2U That at least some of these transliterations were subsequently replaced by translations
can still be seen from several doublets in the manuscripts of the LXX consisting of a
transliteration and a translation, ¢.g. Gen 22:13 7202 - év dutu) oaBex. Cf. Tov, “Loan-
words”™” for further examples; Wutz, Transkriptionen, 54 ff.; Seeligmann, Isaiah, 59.
Transliterations of technical terms are, of course, not covered by this explanation.
Their raison d’étre differs notably from that of the transliterations of unknown words: the
translator either had difficulty finding an appropriate word in the target language or the
Hebrew term was so well known to his (Jewish) readers that a translation would have
detracted from the quality of his version. Our explanation of the transliterations of
506 CHAPTER THIRTY-THREE
developed linguistic sensitivity on the part of the reviser. But in view
of kaige-Th’s root-linked renditions and scrupulous distinctions between
translations (cf. Barthélemy, Devanciers, passim), it is not difficult to
attribute to him the requisite sophistication.
The above description is not meant to imply that all revisers left
unknown words untranslated, or that at the source of the transliteration
of a difficult word there always lies a revision: The reviser of Reigns yd
(kaige) and Theodotion did not invent this practice, for the technique
had been used previously in the Old Greek translation.
The practice of leaving unknown words untranslated has been shown
to be characteristic of kaige in Reigns y& and Theodotion (i.e. the notes
referring to the contents of Origen’s sixth column). It is probable that
Barthélemy is correct in assuming that the two2? are identical, or,
rather, belonged to the same revisional school. O’Connell, Exodus,
points in the same direction. If this opinion can be corroborated by
further evidence, the practice discussed here can be seen not only as a
common feature of two different revisers, but as a characteristic element
of one and the same revisional school. When used critically, this
criterion may also be applied to other members of the same group.??
External evidence corroborates our findings: Theodotion, with whom
kaige-Th was connected in antiquity, is said (by Jerome, see n. 4) to
have frequently used this practice.
If this line of argument can be sustained, the consensus about
Theodotion’s transliterations founded on Field’s work is intrinsically
correct, although a reformulation, based upon different evidence, is
required.
unknown words (the majority in the LXX and Theodotion) is not affected by that of the
Era nsHleia ons of technical terms. The two groups have a different background.
2 Le. kaige-Reigns y5 and the quotations from Origen’s sixth column except for those
books in which “0” clearly does not refer to kaige-Th, viz. 2 Sam 11:2-1—Kgs 2:11 and
Dodekapropheton (see Barthélemy, Devanciers, 128-36, 253-60 and Cross, “Biblical Text,”
290):
23 Using Barthélemy’s list of members of this group, we have found transliterations of
unknown words in Theodotion in Job and Jeremiah (both in asterized and non-asterized
passages), Th-Dan (since Theodotion and Th-Dan are not to be identified as one reviser,
the kaige-like revision 8HevXIIgr also contains one transliteration of an unknown word
(o-102 in Zeph 1:4 = Th ad loc. = kaige 2 Kgs 23:5). It should, however, be stressed that
transliterations do not necessarily point to kaige-Th and that the lack of transliterations
cannot exclude a certain section from belonging to kaige-Th.
TRANSLITERATIONS 507
APPENDIX
1. Transliterations in the LXX
The following list, based on HR,*4 contains transliterated words that
were apparently unknown to the translators of the LXX. The majority of
the words in the list are hapax legomena or rare words in the Bible
(sometimes rare in a certain book). Some prefixed articles and suffixed
elements appear to be included in the transliteration. Transliterations
which are components of doublets (see n. 20) are denoted ‘d.’
a. Words left untranslated apparently because they were unknown
Gen 22:13 es,
Judg 5:7 A... pmo
Judg 5:16 A... o-nown (contrast v. 15 A)
judgo27 A. nin
Judg 6:26 nya
Judg 8:7, 16 o7372
Judg 8:26 A... mou) d.
ibid. onnw
Judg 9:27 B... myn
1 Sam 5:4 jnoan d.
1 Sam 6:8, 11, 15
Lsam 146,.11,.12,15
TANT ,TIN3
(7)aNM (v. 8) d.
1 Sam 14:25 Bi
1 Sam 20:19, 41 *319Kn2 dd.
1 Sam 21:8(7) TV)
T-Sam 30'3).15, 15,23 TW (LXX: 1773)
2 Sami 17:19 mow
2 Sam 17:29 now
1 Kgs 5:25 (11) n?9n (LXX: paxetp or sim.)
1 Kgs 14:28 NN
1 Kgs 19:4 on
2 Kgs 2:14 NWT AN
2 Kgs 3:4 13
24 The evidence is not exhaustive, if only because HR does not list transliterations
reconstructed from Hellenized forms, ¢.g. Jer 31(38):21 O°0N - TLULpwpLL Spohn] ttywptav
codd. gr. et verss.
The word is reconstructed as *20x by Driver, Samuel, 167-168 according to the form of
the Greek transliteration (epyaf, apyaB). MT has 3x7 in 20:19 and 2237 in 20:41.
508
CHAPTER THIRTY-THREE
2 Kgs 4:39 nN
(2 Kgs 4:42 Ax 1)2?82)
2 Kgs 8:15 129” (cf. n. 46)
2 Kgs 9:13 07}
2 Kgs 10:10 NIDN
2 Kgs 10:22 mnn?n d. (LXX: peo@aad or sim.)
2 Kgs 11:8, 15 ni7w(7)
2 Kgs 11:12 Bie
2 Kgs 12 passim; 22:5, 6 (Qa)?
2 Kgs 15:5 mvonn
2 Kgs 20:13 mn)
2 Kgs 23:5 o-1n>
ibid. nym
2 Kgs 23:7 aena (LXX: orn> (?); cf. n. 46)
2 Kgs 25:12 on.” (LXX: 0723)
2 Kgs 25:14 B... ay?
2 Kgs 25:17 ter nnd (LXX: xw6ap)
ibid. Daw
1 Chr 12:23 77) (LXX: 7773)
LChr 15:20 mnyy 9y
iehri5.21 mprnwn oy
1 Chr 21:20 o-xannne?
1 Chr 26:15, 17 ODOR
1 Chr 28:11 (20)
yor (LXX: Cakyw avtod)
1 Chr 28:17 A... "F1DD
1 Chr 29:2 Onw
2 Chr 3:16 nnwiw
2Chr 12-13 ny}
ibid. nin
2 Chr 25:18 bis monn
2 Chr 26:21 (Q) mwonn
Esdr 1:9(10); 8:27 “7195
Esdr 2:62 oowneny
Neh 1:1; 7:2 aaa)
Cant 4:4 nio2n
Cant 4:14 mDAN
Cant 5:11 15
26 Cf. 2 Kgs 12:8(9) B&éAvyya B] AM omnes Bedex.
2? The translator of 1 Chr 21:20 possibly knew the meaning of the verb, but the syntax of
his translation was so completely different from that of MT that he was unable to translate
the verb adequately and consequently left it untranslated.
Amos 1:1
Isa 5:2
Isa 39:2
Jer 7:18; 44(51):19
Jer 8:7
ibid.
Jer 31(38):21
ibid.
Jer 37(44):16
Jer 46(26):15
Jer 52:19
Jer ibid.
(Ezek 1:14 A...
Ezek 27:16
ibid.
Ezek 40, 41 passim
Ezek 40 passim
Ezek 41:8
TRANSLITERATIONS 509
rap (LAX)
Ww
mn)
‘oss,
a8
mon
n7yy29
onion (cf. n. 24)
nvin
nnoi (LXX: €buyev 6° Atis)
mao (LXX: caddwé; MT: 075d)
minm (LXX: pacpapw; MT: ny)
p12)
qaaa (1EXXKS 9595)
NANT
DR
Nn
ern (LXX: TO Opaed[?])
b. Translations of unknown or difficult words probably understood as
proper nouns?!
Gen 15:2
Gen 35:16; 48:7
Gen 36:24
Gen 50:10, 11
Deut 3:5
Josh 11:2; 12:23; 17:11
Judg 1:19
Judg 20:43 B...
Judg 20:48 B...
1 Sam 6:18
1 Sam 15:3
1 Sam 15:8
> W7
nD
on
TORN
119
nai, 91
19782
rim (LXCX: 17337)
ans
719
onaqnn
oan d.
28 All manuscnipts read aypov doubtlessly resulting from ayoup, cf. Ziegler, Bettrage, 84.
29 See J, Ziegler, Beitrage, 85.
30}. Ziegler, Beitrige, 86.
These examples should not be explained as reflecting different interpretations of the
Hebrew. In all these instances the word or immediate context caused difficulties to the
translator who, therefore, avoided the problem by representing the Hebrew as a proper
noun.
2
237 was transliterated as PnyaB because of contextual difficulties.
510 CHAPTER THIRTY-THREE
1 Sam 20:20 7107(7)
1 Sam 21:3 "NPN 72D dd.
1 Sam 23:14, 19: 24:22 Tyo ,mitxns d.
1 Sam 24:2 ae2u"
2 Saino 12 innn, Q: nnn (LXX: 6atraL)
2 Sam 15:32; 16:1 wx (Summit)
2 Sam 21:20 Tm, Q: 770
1 Kgs 4:11 nD}
1 Kgs 15:22 ‘PI TPR
2 Kgs 5:19 aie)
2 Kgs 6:8 “NPR
2 Kgs 12:10(9) para nan
2 Kgs 15:10 B... py 929
2 Kgs 23:11 on
Neh 3:8 age
Ps 74(73):15 rable:
Zech 14:10 TANT
Jer 21:13 WS
Jer 48(31):1, 39 nnn
Dan 8:13 1D
1 Chr 4:22 apa ate eae Oe
2. Theodotion
a. Transliterations of probably unknown words transmitted as Th°®
The following list is based on the data provided by Field and is
compiled in accordance with the principles described in the beginning
of this study.”
33 4 Sam 23:14 Maoepep B...; 19 B... Neooapa, A... Meoapa; 24:23 B... Mecoapa.
The translator transliterated because of contextual difficulties. While B... have a
transliteration of }"2°2, manuscripts A... contain only a transliteration of navn.
5 Manuscripts bore, teB\aay = 0992"?
Theodotion’s readings are generally transmitted piecemeal, which leaves us in the
dark as to his rendering of the whole context. As a result, one ts unable to ascertain
whether some difficult words, when transliterated, were considered proper nouns, as in
some cases in the LXX (see list 1b above). However, the words that were transliterated by
Theodotion, when viewed in their Hebrew context, do not seem to allow for the inclusion
of such a sub-group b.
37 As the fifth column of Origen’s Hexapla in Reigns By contains kaige-Th, Theodotion’s
revision was probably not found in its usual place, viz. the sixth column. Four
transliterations are listed for that column. The same Hebrew words were transliterated also
by kaige-Th (= ‘LXX’). However, the Hebrew basis of two of them differed from that of
TRANSLITERATIONS 511
Lev 13:6 nndon
Lev 18:23 Dan
1 Sam 19:13 a al
1 Sam 26:7 pphina)
2 Kgs 8:15 919”
2 Kgs 9:13 og).
2 Kgs 11:12 ale
2 Kgs 23:7 ona
Job 8:11 ats
Job 36:30 (sub ®)
Job 37:12 (sub @)
Mex (Th :177x)
inaianna, Q: rn’?anna (Th: év
Bee BovrAaddw8)
Job 39:13 (sub ®) Ton
Am 1:1 O77}
Zeph 1:4 o>
Isa 2:20 miqatan (MT: nip 75n); Th: dapdapwé)
Isa 3:24 71°nD
Isa 17:9 wan
ibid. mals:
Isa 19:15 TDD
Isa ibid. 71018
Isa 22:15 ]D0
Isa 22:24 (sub @) NWN
Isa 23:13 (sub ®) on
Isa 38:14 o10 (Th: 070)
ibid. WW
Isa 41:19 (sub ®); 60:13 WD
ibid. WWRN
Isa 43:20 on
Isa 47:2 7NnX
Isa 63:3 m5
Isa 64:5(6) oy
Jer 8:7 Ton
Jer 29(36):17 (sub @) oyvy
Jer 38(45):12 (sub ®) on
Jer 52:16 alah oO ayanm gaan
kaige-Th: 2 Kgs 23:7 kaige-Th xetTveip(v) = orn> (?), sixth column BeO6cetp = MT arna. 2
Kgs 8:15 kaige-Th xaBfa B, aBpa A (=?), sixth column paxBap = MT 123%. The evidence is
too scanty to be decisive. However, it appears that the sixth column of Reigns 5, similar to
its counterpart in Reigns By, is not identical to kaige-Th of the same section.
512 CHAPTER THIRTY-THREE
Ezek 9:2, 11 nop
Ezek 16:10, 13 "wn
Ezek 27:16 791 (Th: adek)
Ezek 27:19 1,2
Ezek 27:24 ov?%2n (Th: payadty)
ibid. ov) (Th: yadtua)
ibid. on
Ezek 27:27 273
Ezek 41:13; 42:1 algie
Ezek 44:18 yr
Ezek 46:17 aa
The following transliterations were not mentioned by Field:
Job 28:18 (sub ®) a)
Job 38:32 (sub ®) ny
Job 39:13 (sub @) miopysa)
Job ibid. (sub ®) 7]
Isa 3:24 MwWPN
Isa 19:7 nw
Jer 44(51):19 (sub ®) maxyun? (Th: kal coBBa)
Ezek 9:2, 11° 10:2 ‘aha
Ezek 27:17 119
Ezek 30:9 (sub ®) ale
b. Transliterations of unknown words in Th-Daniel
3 O7ann
4:10, 14, 20 (13, 17, 23) a ie,
8:2,3,6 ITN
8:13 90D
10255 226.7 alin ba
re Te: aban tra)
L14S VITOR
c. Transliterations of unknown words, transmitted as collective
readings?
Job 30:4 mama’ 6
Isa 5:2 pwa’ &
Isa 13:21; 34:14 Be Oki oO 40b XS)
Isa 40:15 Pla’ §’
38 Cf n. 7.
CHAPTER THIRTY-FOUR
UNE INSCRIPTION GRECQUE D’ORIGINE SAMARITAINE
TROUVEE A THESSALONIQUE
En 1968 B. Lifschitz et J. Schiby (L-5) ont publié une inscription
samaritaine en grec et en hébreu, trouvée dans une synagogue de
Thessalonique et composée des trois sections suivantes:!
1. Deux bénédictions hébraiques écrites en caracteres samaritains
(lignes: letl>);
2. La ‘bénédiction des prétres’ (Num 6:22-27) en grec (lignes 2-14);
3. Une dédicace en grec, par un certain Siricius (lignes 16-20).
Nous nous proposons de discuter le caractére du texte biblique de cette
inscription, au-dela des remarques faites par les éditeurs. Pour ce faire,
nour reproduirons d’abord en colonnes paralléles le texte de l’inscrip-
tion, tel qui‘il a été imprimé par L-S,2 le texte du codex B, accompagné
des variantes, et le texte massorétique (ITM). L’ordre des versets dans B,
tel qu'il est présenté ci-dessous, a été modifié pour s’accorder a la fois a
celui du TM et a celui de I’inscription (voir ci-dessous).
Inscription LXx8 TM Num 6:22-27
Kal €AdAnoev K(UpLO)S Kal EAAANOEV KUPLOS mim Tate2
weTa Mouor Aéywv TpOs Muvay AéE-ywv TN? TWD 28
AdAnaov TH * Aapuv \dAr}Gov4 Aapwv AX 2x 79723
Kal Tots uLots ab[tot] Kal tots uLots avtot 712 ONI
Eyuv hE yuwv TIN?
oUTuws ebAoyraleTe | OUTWS EVAOYNOETE wei s
1 “Une synagogue samaritaine a Thessalonique,” RB 75 (1968) 368-378. L’inscription
avait été antérieurement publiée sans aucun commentaire par S. Pélékidis dans
Tletpayyéva tov 6 &teQvovs BuCavTtvoroyikov Luvedprovd (Becaadovlkn, 12-19’ AttpiAlou
1953), I (Athens 1955) 408. Subséquemment, Lifschitz a de nouveau publié l’inscription
sans aucun changement dans son “Prolegomenon” a J.-B. Frey, Corpus of Jewish
Inscriptions—Jewish Inscriptions from the Third Century B.C. to the Seventh Century A.D. (1936;
repr. New York 1975) 70-75.
Pour la disposition précise de l‘inscription, le lecteur est renvoyé a LSJ. Nous divisons
le texte en petites unités pour faciliter son étude.
514
Tous ulLovs *I (cpa)nr
CHAPTER THIRTY-FOUR
Tous uLovs Iapanr
5x1Ww? 3D NN
ettate avtots héyovtes adToisP on? 7X
EVAOYHOEL GE K(UPLO)S evAOyHaaL® oe KUPLOS mia yoa°24
Kal PvAGEEL OE Kal dvddEa4 Jnnw
éEtmidavet K(UpLO)S éTLpbavar® KUpLOS TT N29
TO TPSOWTIOV AUTO TO TPdOWTIOV aUTOU 715
pds GE éml oe wis:
Kal dyatmoeL Oe Kal édenoaf oe aE ighe
étrapet K(UPLO)S ETapat8& KUPLOS TT Nwyr26
TO TPOOWTOV avT(o)d TO TpdOWTIOV aVTOU yp
TIpdS GE éml oe POR
Kal to.joet cot el(p4)vnv kal Susy cot elphynv ow 72 ow
Kal O6joeTar Kal é€mLO4covoLV wy?
TO 6VOLE [Lou TO OVOULA [LOU "aw NN
étl Tots ulLots "I(opayhd — EtIL TOUS ULots* Iopanr Paw? ID DY
KaYW EVAD'YHOW AUTOUS Kal éya) KUpLos! ODN 738)
EVAOYT|OW alTous
Apparatus criticus
a bw+Tu Bo Co Chr étdavet
e] TPds aUTOUS f hBoCo é€defoer
c n*BoCoChr evdoyroet Chr evdAoyHoeu
o evdoyNoT Bo Co Chr étrapet
d texte Bap |] Bo Co Chr dvudd€eu oe F év tots vlots
AFGHMN rell Arm Co Eth La Cyr dg éml tots uvlots
Thdt + ae i <46> Ethc°om
La comparaison entre ces textes ainsi mis en paralléle appelle les
remarques suivantes qui portent d’abord sur les ressemblances (accords)
entre les deux textes grecs, puis sur leurs différences (désaccords).
a og
On doit noter les accords suivants entre ‘inscription et LXxB:
1. Dans les deux textes le tétragramme est représenté cing fois par
KUpLOS (vv. 22, 24, 25, 26, 27). Dans une discussion concernant une source
grecque, qui est peut-étre d’origine samaritaine (P. Giessen 13, 19, 22 et
26 contenant des fragments du Deuteronome), nous avons essayé de
montrer que l’usage de cet équivalent pour le tétragramme n’a pas pu
avoir son origine chez les Samaritains, mais qu’il est basé sur le vocabu-
laire des LXX (cf. Tov, “Pap. Giessen”*).
2. La traduction de 1x* du v. 25 par émidalvw dans les deux textes ne
peut pas étre une coincidence. Cette traduction identique vient
INSCRIPTION GRECQUE 515
probablement des LXX ot. TxA est habituellement rendu par (€T-)
gatvu.?
3. Au v. 26, le verbe xwi (de la locution 0715 Xw*) est rendu par
étidatvw. Dans les LXX, les équivalents les plus fréquents de xwi sont,
par ordre de fréquence, alpw, A\ayBavw, dvarayBavw, et ETalpw y compris
dans la locution en question et dans l’expression similaire Oy XW1.
4. Dans les vv. 22 et 23, 1px est rendu par \éywv et non pas, par
example, par l’infinitif du verbe.
En raison de ces accords, si peu nombreux soient-ils,> il parait
probable que les deux textes aient une base commune; c’est l’indice que
cette inscription fait partie de la tradition des LXX. Cette supposition
peut étre renforcée par deux accords supplémentaires, bien que moins
significatifs:
113 - ebdAoyéw (vv. 23, 24, 27), et non pas un verbe parallele;
nw - durdoow (v. 24), ie. le verbe simple et non pas l’un de ses
composes.
Nous passons maintenant aux différences entre les deux textes. Si nous
avons raison de supposer que I’inscription de Thessalonique se rattache
a la tradition des LXX, les différences entre les deux doivent étre le
résultat de révisions et sont ainsi les témoins de cette activité de
révision, a l’intérieur de la tradition textuelle des LXX. Il est suggéré
que l’inscription de Thessalonique, comme les Pap. Giessen, contient une
révision tendant a représenter l’hébreu d’une maniére plus précise.
L’argumentation utilisée sera la méme que celle mise en oeuvre au sujet
des P. Giessen (cf. Tov, “Pap. Giessen”*; TCHB, 143-148).
a. L’inscription présente un texte révisé des LXX, plus proche d'un texte
hébreu, parfois different du TM
v. 23 JAnX ON - *Aapuv LXX ] TH ’ Aapoiv inscr. (= mss bw des LXX).
L’addition de I’article To est a interpréter comme représentant bx.6
3 Le sens du verbe émalvw contient un élément de brillance; d’aprés les dictionnaires
cependant, ce sens est plus manifeste dans le verbe simple. En conséquence, il n’y a pas
lieu de supposer que emidalvw refléte des formes de la racine 1X9.
4 CF. 2 Sam 2:22 (alpw) et Job 11:15 (dvahapBdvw).
Dans des textes plus longs, la base commune a deux traditions doit étre établie a partir
d’un plus grand nombre d’accords. Mais dans cette petite inscription on ne doit pas
s’attendre a de nombreux accords.
6 Dans les constructions semblables a celle-ci (AdAnoov’ Aapuy) l’article est rarement omis
dans les LXX devant le nom propre (voir cependant Exod 6:11 et Lev 6:18 [25]).
516 CHAPTER THIRTY-FOUR
v. 23 710n - Méyovtes LXX ] eltate inscr. reflétant peut-étre nx.”
vv. 24-26 J272", Taw, TW", 73771, Xv, ow. En traduisant ces jussifs, le
traducteur® des LXX a utilisé des optatifs aoristes alors que l’inscrip-
tion a des indicatifs futurs. Du point de vue formel, l’indicatif futur est
une représentation plus stéréotypée de l’hébreu.
v. 25 qin) - kal €dXerjoat ce LXX | kal dyatmoet oe inscr. L’équivalent
dans les LXX, qui s’explique par l’usage des LXX en ce qui concerne la
racine 73n,? ne rend pas bien le sens de |’hébreu. En conséquence, éde€w
(LXX) est remplacé par dyatdw. Lifschitz, “Prolegomenon” (cf. n. 1) 74
justifie la possibilité que dyatdw soit linguistiquement approprié.
v.27 wi - kal ETLOHWoovoLV LXX | kal @joetat inscr. Les LXX reflétent
le TM. Au contraire, la lecon utilisée par l’inscription peut refléter une
variante “nw n& aw (1). Elle pourrait aussi bien étre le résultat d’une
modification du traducteur grec: substitution d’une forme passive a une
forme active du verbe, pour faciliter le passage du v. 26 au v. 27.
v. 27 kiptog LXX ] Comme le TM, I’inscription omet ce mot.!9
On peut encore ajouter les deux points suivants:
L’ordre des versets dans les LXX est: 22, 23, 25, 26, 27, 24. L’inscrip-
tion, au contraire, reproduit l’ordre du TM, ordre qui se trouve aussi dans
plusieurs témoins des LXX qui ont été harmonisés avec le TM: Gckquxy?’
BoW Co Eth‘. L’ordre des versets dans l’inscription s’expli-que, soit par
une révision destinée a plus de conformité au TM, soit parce qu’elle
dépend d'une tradition attestée par les teémoins mentionnés ci-dessus. La
premiére possibilité semble probable, mais l’autre ne peut étre ni
prouvé ni réfutée.
v. 24 JFrinw - Kal durdEat Bag |] kai dvdAdEet ce inscr. La lecon de
l‘inscription, conforme a la majorité des témoins des LXX, peut étre
7 Cf. d'autres traductions anciennes qui utilisent aussi un impératif ou une forme
similaire.
D’aprés Brooke-McLean, quelques manuscrits des LXX offrent des paralléles avec cette
inscription; voir les variantes indiquées au-dessous de notre tableau synoptique.
? Dans la Bible les formes de Ia racine jin signifient a la fois “étre gracieux’ et ‘avoir pitié.’
Pour ce dernier sens, éhe€w et olktlpw sont les équivalents normaux dans les LXX (p. ex. Ps
37 (38):21,26; Prov 14:31). Toutefois, ces équivalents sont employés d’une maniére impropre
pour tous les sens de }1n: ainsi par exemple dans Gen 33:5 ot: hAénoev rend mal le sens de
I"hébreu, comme dans le verset que nous examinons ou I’impropriété est peut-€tre moins
patente. Des traductions similaires se retrouvent tout au long des versions araméennes
(ann, din, 7m).
0 L’addition de kipiog dans les LXX a pu étre causée par le contexte. Dans ces versets,
Dieu apparait quatre fois a la troiséme personne, alors que Dieu a la premiére personne est
utilisé seulement ict: le traducteur a peut-étre voulu insister sur le fait que c’est bien Dieu
qui est le sujet du verbe ‘bénir.’ Voir cependant Frankel, Einfluss, 178, quia suggéré que ce
kUptog additionel refléte un tetragramme abrégé ’* dont l’origine serait, dans ce verset, une
dittographie de la finale du mot précédent 714.
INSCRIPTION GRECQUE ay
basée soit sur ces manuscrits, soit sur le texte plus court attesté par Baz
auquel cas le pronom aurait étre ajouté dans l’inscription. La lecon courte
semble plus originale, car le traducteur aurait probablement considéré
comme inutile et inélégant de répéter le pronom complément oe apres
evAoyroat ae KUpLos Kal puddéEat.
b. L'inscription remplace des mots des LXX par des synonymes
Les réviseurs des LXX ont souvent remplacés des mots par d’autres qu’ils
ont trouvé préférables, pour une raison ou une autre. Pour nous, ces
traductions modifiées peuvent ne pas toujours apparaitre comme des
reproductions plus fidéles de l’hébreu, mais le réviseur avait
probablement ses objectifs propres. Voici trois exemples de ces révisions-
substitutions:
v. 22 Ox - tpd¢ LXX ] peta inscr.!!
v.25 Jon - Et o€ LXX ] pds o€ inscr.
v. 27 ow) - kal Bain LXX ] kat Totjoer inscr. Ces deux verbes grecques
servent aussi a rendre le méme verbe hébreu dans d’autres passages des
LXX.
En conclusion, on peut suggérer que l’inscription samaritaine trouvée a
Thessalonique fait partie de la tradition des LXX et qu’elle représente
une révision de cette tradition.!* La situation et le caractére que nous
croyons devoir reconnaitre a cette inscription sont donc assez
comparables a ceux des Pap. Giessen (cf. Tov, “Pap. Giessen”*).}8
11 Dans les LXX, \adhéw Lretd traduit habituellement nx 127 / OY 737 et seulement une
fois Ox 327 (Ezek 3:10). Cf. Helbing, Kasussyntax, 239.
2 Lifschitz, “Prolegomenon” {voir n. 1) 74 a également souligné la fidélité a |’hébreu de
l‘inscription. Dans cette seconde édition de l’inscription, Lifschitz (1975) a charactérisé
l‘inscription comme provenant d’une tradition indépendante des LXX. D’aprés Lifschitz,
l’inscription réfléte une traduction samaritaine indépendante de la Bible hébreu. Mon
article original (1974) n’est pas cité par Lifschitz.
Dans l'état actuel de Ja recherche, il est impossible de déterminer Ia relation précise de
notre inscription avec ce qu’on appelle le Hapape.tikdy, parce que la nature exacte de ce
demier n’a pas encore été établie. L’un de ses traits caractéristiques a cependant été bien
illustré par S. Kohn, MGWJ 38 (1893-1894) 1-7, 49-67 qui a montré que plusieurs des
fragments du Zapapettikdv dérivent du T>@™ et non de leur texte hébreu. Selon ce critére,
l’inscription de Thessalonique ne devrait pas étre considerée comme paralléle au
Yapapetixdv: les éditions existantes du T°2™ montrent plusieurs différences entre Team
et l’inscription de Thessalonique.
CHAPTER THIRTY-FIVE
THE RELATION BETWEEN THE GREEK VERSIONS OF BARUCH
AND DANIEL
As early as 1879, J.J. Kneucker provided detailed proof that at least the
first part of Baruch (1:1-3:8) was originally written in Hebrew.!
Kneucker also published his own reconstruction of the original text of
the whole of Baruch.*
Several scholars recognized that the author of the Hebrew book of
Baruch frequently quoted from the books of Jeremiah and Daniel. The
Greek text of Baruch is also closely related to the Greek texts of
Jeremiah and Daniel. This close relationship between the Greek texts
of Baruch and Jeremiah has been explained in various ways, ranging
from imitation of the language of Jeremiah by the Greek translator of
Baruch to identity of translators. The present paper focuses on the
relation between the two Greek texts of Daniel and the Greek
translation of Baruch.
Various scholars have pointed out that the Hebrew Vorlage of the
prayer of repentance in Bar 1:15-2:19 is an elaboration of Dan 9:5-19.° It
has also been recognized that the Greek translation of Baruch is closely
related to one of the two translations of Daniel, that of Th(eodotion),
because in several instances the latter concurs with Baruch against
Daniel-LXX.4 These agreements between Baruch and Daniel-Th were
1 Das Buch Baruch (Leipzig 1879) 20 ff.
2 Pp. 351 ff. See also the present author’s reconstruction The Book of Baruch Also Called I
Baruch (Greek and Hebrew) (TS 8, Pseudepigrapha Series 6; Missoula, MT 1975).
3 See, e.g., Eissfeldt, Introduction, 593 and especially B.N. Wambacq, “Les priéres de
Baruch (1,15-2,19) et de Daniel (9,5-19),” Brb 40 (1959) 463-475. The agreement between
the two texts is so striking (see Wambacq, ibid., in detail), that immediate literary
dependence is highly probable. With Wambacq we assume that the Hebrew Vorlage of
Baruch elaborated upon Daniel, but the reverse assumption is not impossible. For our
purpose the exact direction of dependence need not be determined. It is not impossible
that Baruch relied on a text similar to Daniel, but such a text has not been preserved.
J. Gwynn, “Theodotion” in W. Smith and H. Wace (eds.), A Dictionary of Christian
Biography (London 1887) IV, 976, was the first scholar to draw attention to Baruch’s
520 CHAPTER THIRTY-FIVE
taken by scholars as additional proof of the existence of a so-called
‘proto-Theodotion’ translation: the historical Theodotion presumably
lived in the second century CE—certainly after the time of the
composition of Daniel and Baruch as well as their Greek translations—
but scholars have discovered several quotations of “‘Theodotion’ from
Daniel and other books of the Bible in sources antedating the time of
the historical Theodotion.°
The proto-Theodotionic problem has been much discussed in past
generations, and ever since the 1963 publication of Barthélemy,
Devanciers that discussion has been revived. It is now evident that the
enigmatic proto-Theodotion is none other than the kaige-Th revision,
which in antiquity was ascribed to the historical Theodotion, and
probably preceded the turn of the era.®
The present study attempts to show that some agreements exist
between Baruch and Daniel-Th (differing from Daniel-LXX), but most
(if not all) of them have no bearing upon the nature of the relationship
between Baruch and Th-Daniel and are, in a way, coincidental. Thus
the below-mentioned agreements cannot be taken as proof of the
existence of a proto-Theodotionic translation or, for that matter, of
Baruch’s use of kaige-Th. This shows that in unraveling the web of
relationships between different translations one should beware not to
determine interdependence on the basis of superficial resemblances
only.
In order to clarify the relationship between the translations of
Baruch, Daniel-LXX and Daniel-Th, their renderings of the same
dependence upon Daniel-Th. He was followed by T. André, Les Apocryphes de l’Ancien
Testament (Florence 1903) 251; E. Schtrer, ThLZ 29 (1904) 254-256; O.C. Whitehouse apud
R.H. Charles (ed.), The Apocrypha and Pseudepigrapha of the Old Testament (Oxford 1913) I,
574; Swete, Introduction, 48; H.St.J. Thackeray apud C. Gore, A New Commentary on Holy
Scripture (London 1928) II, 105; Jellicoe, SMS, 88, 93; A. Bludau, “Die Apocalypse und
Theodotions Danieltibersetzung,” Theologische Quartalschrift 79 (1897) 18 was undecided
whether Baruch depended upon Daniel-Th or vice versa. A similar opinion has been
expressed by Schmitt, Theodotion, 13-14.
In contrast to these scholars, Kneucker, Baruch, 82 felt that the two translations were
independent. However, Kneucker did not submit the problem to a detailed investigation.
Nor did J.A. Montgomery, A Critical and Exegetical Commentary on the Book of Daniel
(Edinburgh 1927) 50, who expressed a similar opinion. However, his view is based upon
incorrect arguments, since nearly all the differences he noted resulted from different
Vorlagen (see n. 7, below).
> For a convenient summary of the evidence relating to proto-Theodotion, see Schmitt,
Theodotion, 11 ff.
6 See Barthélemy, Devanciers, 167 and Cross, ALQ, 171, n. 13.
BARUCH AND DANIEL opal
Hebrew word are classified in the analysis below.” In the course of this
classification, only major variants are noted. In the examples to be
mentioned below, our reconstruction differs only once from that of
Kneucker.®
1. Baruch = Daniel-LXX # Dantel-Th?
lf vow Bar 1:18 = Dan-LXX 9:10 aKOUW
Dan-Th 9:10 eloakovin!9
2 Mina: -Bar'2:2:. = Dan-LxXxX 9:13 KaTa Ta ye ypayweéva
Dan-Th 9:13 KaQus yé ypatTTar
2: Nx71p1 Bar 2:15 = Dan-LXX 9:18 étreKAH On}!
Dan-Th 9:18 € TLKEKATTAL
Neither the agreements nor the differences are of special importance
since the synonymous renderings!* recur passim in the LXX.
2. Baruch = Dantel-Th # Daniel-LXX
a. Synonymous renderings
4, nin Bar 2:2 = Dan-Th 9:12 UTOKATW
Dan-LXX uTTO
5: qpw Bar 2:9 = Dan-Th 9:14 Yonyopew
Dan-LXX aYypuTIVe wW
6. ynw Bar 2:14 = Dan-Th 9:17 elLoaKkouw
Dan-LXX éTaKOUW
yi vow Bar 2:16 = Dan-Th 9:18 aKkouw
Dan-LXX € TaKOUW
7 Agreements or disagreements which resulted from differences in Vorlagen, both extant
and assumed (e.g., Bar 2:14 = Dan-Th 9:17 €vexév cov = Dan-LXX Evekev Tuiv B0UdAWY dou)
belong to a different level of discussion.
8 In item (2) Kneucker reconstructed 31ND WWRrR?2.
7 79 is uniformly rendered by tais in Baruch (4 x) and Dantel-LXX (11 x). Daniel-Th, on
the other hand, most likely distinguished between the religious use of 7112Y (= S00A0¢ 6x;
10:17 tats), and the non-religious use (tats 5 x). This example is not listed since the
Hebrew of Baruch does not contain examples of the non-religious use of 71.
0 For additional examples of the interchange between the simplex of dkovw and this
ee see Helbing, Kassussyntax, 154.
: Manuscripts A... of Baruch read é€tLkékAnrTat.
Two renderings are ‘synonymous’ when they are found concurrently in the LXX,
without any grammatical or contextual condition determining the occurrence of one
rather than the other. Many of these renderings appear elsewhere as variants in the
manuscripts of the LXX. Generally the Greek words themselves are also synonymous.
522 CHAPTER THIRTY-FIVE
Neither the agreements nor the differences are of special importance
since the synonymous renderings recur passim in the LXX.
b. Baruch = Daniel-Th # Daniel-LXX (rare or hapax renderings)
8. (amor) Bar 1:15; = Dan-Th9:7,15 as (h thpépa ath)
2701 Dan-LXX 9:7, 15 Kata (Thy tpépav
TauTny)
The kaph in 714 01(7)D is in the LXX rendered almost exclusively by ws
and rarely by other prepositions.}2
9. vw Bar 1:16
Dan-Th 9:8 doywv (9:6, 8; 11:5)
Dan-LXX 9:8 8uvdotns—limited
to Daniel
Dan-Th 9:10 TOPE VOLLAL
Dan-LXX 9:10 katakod\ov8éw—a
hapax rendering
and a very rare
word in the LXX (2
x in the canonical
books and 3 x in the
deuterocanonical
books).
alee Vili) ‘Bare. Dan-Th 9:12 Adyou (= K)
©O sat Aoyos (=Q) Dan-LXX 9:12 tTpooTdypata—an
uncommon equiv-
alent, occurring
especially in Dan-
LXX.
Dan-Th 9:13 VO|LOS
Dan-LXX 9:13 8vaO4Kn-a hapax
rendering in the
LXX.
Dan-Th 9:13 Séouat
Dan-LXX 9:13 ék€ntTéw—a hapax
rendering in the
LXX.
10. 354 Bar 1:18
12; mon: Bar 2:2
13. mn Bar 2:8
13 Womep (1 x), doel (1 x), KaBus (2 x), KaTd (1 x).
BARUCH AND DANIEL 523
14. 178) «Bar 2:11, = Dan-Th 9:15, kupLe
12, 14 16,17
Dan-LXX 9:15, 6Sé€otrota—elsewhere
16, 17 five times in the
LXxX.
Daniel-LXX presents a very free translation and therefore it causes no
surprise that its renderings mentioned above are very rare or
unparalleled elsewhere in Greek Scripture. On the other hand, the
renderings which are common fo Daniel-Th and Baruch recur passim in
the LXX. Thus the agreements between the two are mere ‘optical
illusions.’
c. Bar = Dan-Th (stereotyped) # Dan-LXX
LS: 199 Bar 1:18 = Dan-Th 9:10 Kata tedowtov
Dan-LXX EvusTriov !4
16. wy ara Bar 2:7 = Dan-Th 9:13 tev €¢’ tds
Dan-LXX ETM AVEV THLtV
17. mex wwe Bar 2:11 = Dan-Th 9:15 ds é&tyayes
Dan-LXX 6 €Eayayuv
18. not Bar 2:16 = Dan-Th 9:18 kdjtvov (...7d obs Gov)
(FUN...) Dan-LXX tmpdaxes (...Td oUS COU)
19. x2 °D Bar 2:19 = Dan-Th 9:18 6t. ovK
Dan-LXX ov ydp
Stereotyped renderings are not exclusively characteristic of Baruch and
Daniel-Th. Therefore they do not indicate any special relationship
between the two versions. Both translators show a tendency towards
literalness which in the cases discussed sets them apart from Daniel-
LXxX.
3. Baruch # Daniel-Th = Dantel-LXX
20. o3970 nw2 11> Bar 1:15 (= Hiv &€ (h) aloxyivn tav
2:619) mpoodtrwv!®
14 For the differences between the Vorlagen of Baruch = Daniel-Th katd tpdowtov uy
and Daniel-LXX évuittov Muoy kal hoy, see n. 7.
15 Bar 2:6 Hutv 6 ... Yaloydvy Tov trpoodtwv.
An investigation of the rendering 071) - tpdowtra in the LXX does not justify the
assumption that the LXX translators used the plural form of tpdowtov to conform with the
ending of o731D. It rather seems that the translators sometimes adapted the form of
524 CHAPTER THIRTY-FIVE
Dan-LXX 9:7. = Dan-Th ‘*ptv (4) aloxytryy toi
MPOOW TOV
ZA. vow Bar 2:1 SLUKAOTHS
Dan-LXX 9:12 = Dan-Th kpims
22. mp1x Bar 2:19 OLKaLWLLATa—very
rare (cf. 2:12, 17).
Dan-LXX 9:18 = Dan-Th Stkatoovvar
23. 71152, Bar: 2:19 KATA TPSOWTTOV GoU
Dan-LXX 9:18 = Dan-Th évatdév cov
Baruch’s rendering is the more stereotyped reflection of the Hebrew. The same
rendering recurs in Bar 1:18 (= Dan-Th 9:10 contra Daniel-LXX) and 2:10.
Neither the agreement between Daniel-Th and Daniel-LXX nor
their disagreement with Baruch are of importance.
4. Baruch # Daniel-Th # Daniel-LXxX
a. General
24. ~y-5Q) Bar 1:15 TOLS KATOLKOUGLY
pow l€povoadny
Dan-LXX 9:7 Kabnpévois ev
le povoaAny
Dan-Th 9:7 Tots évorkoto.v!” év
le€povoaAnp
The three renderings are synonymous. The lack of €v in Baruch follows the
Hebrew./8
25. io «6Bar 1:19 eoxvedidCouev
Dan-LXX 9:11 atTéotnoav
Dan-Th 9:11 éEé€kAtvav
oxeSid¢w occurs only here in the LXX. dglompt and éxkAlva also render this
verb elsewhere in the LXX.
26; mnwy3 Bar 2:2 € TOL ON
Dan-LXX 9:12 eyerron
Dan-Th 9:12 YE'YOVEV
tpdcwtov to the plural noun or pronoun with which it was connected. Thus the plural
form of the noun in Baruch seems to have resulted from harmonization with the plural
hptv.
17 W/L"... KaToLKOUGLV.
18 See Helbing, Kassussyntax, 74-75.
BARUCH AND DANIEL 525
pe minwyi Bar 2:2 é€trolnoey (ETroLH On
AQV...)
Dan-LXX 9:12 éyevn On
Dan-Th 9:12 Ta yevoueva
(yeypappeva A’O™*)
28. Wx Bar 2:2 Kaba
Dan-LXX 9:12 KaB6TL
Dan-Th 9:12 KaTd
b. Baruch shows the same tendency towards literalness as Daniel-Th
29. Tyr wer? Bar 1:15 avOputrw | ovda
Dan-LXX 9:7 dav@puitrots | ovSa
Dan-Th 9:7 — dvdpi LovdSa
30. rwun Bar 2:9 Ta €pya avtTou
Dan-LXX doa dv Totnon
9:14 THY Tolno.y avtot
Dan-Th 9:14
31. wninn oon Bar 2:19 KaTaBdhdopev Tov Edeov Tey
Dan-LXX S5edu€0a Ev Tats TpoGEVYais NuGv
9:18 OLTTOULEY TOV OLKTLOLLOV TL
Dan-Th 9:18
The differences noted between the three translations listed in section a
are not significant. Those mentioned in section b show that Baruch and
Daniel-Th provided different literal renderings, while Daniel-LXX
rendered the Hebrew freely (see especially item 31).
The renderings mentioned above indicate no relationship of
dependence between Baruch and either Daniel-Th or Daniel-LXX. The
differences between Baruch and Daniel-Th = Daniel-LXX (group 3) and
those between all three translations (group 4) are immaterial. One may
also disregard insignificant agreements between Baruch and Daniel-
LXX (group 1) and agreements between Baruch and Daniel-Th when
their rendering is synonymous to Daniel-LXX (group 2a).
Literal renderings (group 2c) probably constitute the key for the
understanding of the relatively numerous agreements between Baruch
and Daniel-Th. None of these is characteristic of Baruch and Daniel-
Th only, and thus no interrelationship between the two need be
postulated. Rather, their shared tendency towards literalness caused
many of their renderings to be identical, as opposed to those of Daniel-
526 CHAPTER THIRTY-FIVE
LXX (groups 2b and 2c). That Baruch and Daniel-Th were two different
literal translators is further corroborated by:
1) group 4b, showing that Baruch and Daniel-Th provide different
literal renderings, while Daniel-LXX rendered the Hebrew freely
(especially item 31);
2) items 23 and 24.
In addition, other data show that both Daniel-Th!? and Bar 1:1—3:8 as
a whole reflect revisions of an original Greek text towards a more
precise representation of the Hebrew. It seems that the resemblances
between Daniel-Th and Baruch are superficial and have no bearing
upon the proto-Theodotionic problem.
19 See Montgomery, Daniel, 46-50 and Schmitt, Daniel, 11-16. Moreover, Danie]-Th
belonged to the kaige-Th recension; cf. Barthélemy, Devanciers, 47; Cross, “History of the
Biblical Text,” 283; P. Grelot, “Les versions grecques de Daniel,” Bib 47 (1966) 381-402.
CHAPTER THIRTY-SIX
THE TEXTUAL HISTORY OF THE SONG OF DEBORAH IN THE A
TEXT OF THE SEPTUAGINT
The main possibilities for explaining the relationship between
manuscripts A... and B... (hereafter: the A text and B text) in Judges
have been explored,! but to date no conclusive evidence has been
produced for any of the suggested theories. One may nevertheless speak
of a common opinion, namely, that the A text is closer to the original
translation than the B text, and that the B text incorporates an early
revision of the original translation. This view has been supported by D.
Barthélemy who included the B text in the kaige-Th group.”
Even if the A text reflects the original translation of Judges, it
should not be considered as reflecting the OG in a pure form because it
contains various doublets as well as interpolations from the B text and
from the Hexapla.? The latter phenomenon is rather frequent,’ but its
correct interpretation must await the analysis of the nature of the B
text.°
! For a summary, see Jellicoe, SMS, 280-283.
Z Barthélemy, Devanciers, 47; see further R.G. Boling, Judges (AB; Garden City, New York
1975) 297-301. Barthélemy’s conclusions are supported by W.R. Bodine, The Greek Text of
[Hats = he sione! Developments (HSM 23; Chico, CA 1980).
See especially J. Schreiner, Septuaginta-Massora des Buches der Richter (Roma 1957)
passim; id., “Textformen und Urtext des Deboraliedes in der Septuaginta,” Bib 42 (1961)
173-200; “Zum B-Text des griechischen Canticum Deborae,” ibid., 333-358.
4 Examples were provided for the Song of Deborah by J. Schreiner, Bib 42 (1961) 346-348;
A.-S. Badillos, “Tradicién griega y texto hebreo de] Canto de Débora (Jue 5),” Sefarad 33
(1973) 251-252; and for the whole book, A. Schulte, De restitutione atque indole genuinae
versionis graecae in libro Judicum (Leipzig 1889) 18-27 (some 200 readings of ‘8’’ in the A
text); O. Pretzl, “Septuagintaprobleme im Buch der Richter,” Bib 7 (1926) 233-269, 353-383;
C.M. Cooper, “Theodotion’s Influence on the Alexandrian Text of Judges,” JBL 67 (1948)
63-68; I. Soisalon-Soininen, Die Textformen der Septuaginta-Ubersetzung des Richterbuches
(AASF B 72,1; Helsinki 1951) 106-109. See further A.V. Billen, “The Hexaplaric Element in
the LXX Version of Judges,” JTS 43 (1942) 12-19.
If the B text reflects kaige-Th, ‘6’ either reflects an alternative tradition of Theodotion
or a completely different text. The same problem obtains with regard to the fifth column
(0’) of the Hexapla. Similar problems with regard to the nature of ‘6’ in the sections
ascribed to kaige-Th in Samuel-Kings have been discussed often.
528 CHAPTER THIRTY-SIX
Within the complicated text history of the Greek texts of Judges, the
text of the Song of Deborah suffered more than any other chapter
probably because of its wide (separate?) diffusion and possible litur-
gical use.© The complicated nature of the textual history of the A text
of Judges is illustrated here through examples of transpositions and
especially of ‘remote doublets.” The number of such phenomena in the A
text is relatively large so that a detailed discussion is in order.
We understand a ‘remote doublet’ to be a double translation, one of
whose elements is transmitted in a completely different place.” The
nature of such remote doublets is illustrated first by v. 14 A...:
ad aos Edpaty ETLULWPHOATO avToUS
B €v KoLAdEL dSeEAMoU Gov Beviautyv Ev Aaots Gov
ba éE €uod Maytp katéBnoav €€epevvasvtes
B Kal €k ZaBovAwy KUPLOS ETTOAELLEL LoL EV BuvaTots
y éxeiOev Ev oKHTTPW EVLOXVOVTOS NyNGEwWs
The Greek translation raises various critical problems vis-a-vis MT; in
those places in which it does not reflect MT, its underlying Hebrew
readings either derived from MT or vice versa:
ETLUWPTNGATO aUTOUS - OwlY MT ow Ww
Ev KOLAGSL ~ pnvi MT onus
ddeddot cou - 7X8 MT PANN
EEE PEVVUVTES = fen MT o-ppnn
éxetOev - avn MT own
Lines baB in the LXX may be translated as following:
a From me, Machir, searchers went down
8B and from Zaboulon God fought for me against (amongst) the mighty
ones.
The meaning of stichs b8 is unclear, but the problem is made easier
when we realize that the words ktpLos étoAé vet Lor €v Suvatots have
been transmitted in the wrong place because these words reflect the
beginning of the previous verse (13) a°71212 >? 77° ‘71. In the translation
doublet in 14b8, 177° is derived by metathesis from the root 17, used in
Syriac as ‘to fight.’ Thus the original text of v. 14b probably read kal é«
© Thus J. Schreiner, Bib 42 (1961) 333 ff.; B. Lindars, JTS N.S. 22 (1971) 5.
7 The two components of a (translation) doublet may occur in juxtaposition or one
element may occur at the end of the sentence, as it were as an appendix, see S. Talmon,
“Double Readings in the Massoretic Text,” Textus 1 (1960) 144-184; Z. Talshir, “Double
Translations in the Septuagint,” in: Cox, VI Congress, 21-63. The uncoordinated occurrence
of the components of the doublets to be mentioned below is rather exceptional.
THE SONG OF DEBORAH 529
ZaBovdAwy éxelbev é€v oxAtmTpw (and in Zaboulon from there with the
sceptre). In this phrase é€xet@ev reflects a variant own corrupted from
o(>)Dwn (note further the variant 02v2 underlying in Sychem of La).
There 1s little doubt that kUptos—S8vvartots reflect 13b o°71212—‘7,
but it cannot be determined why this half-verse was transmitted in the
middle of v. 14. One is therefore left with the assumption of textual
corruption, that is, one component of the translation doublet was
transmitted in the wrong place. These additional words form exactly
one line in codex A® which may have been transposed erroneously as a
whole. V. 14b8 thus forms a second translation of v. 13b with which it
may be compared:
ale Th 3D o-v92
13. KUtpte TaTelvwodv {LoL Tous LaxupoTté pous (Lov)
14 KUpLos E TOAE LEL LLOL Ev Buvatots
Neither translation is an exact replication of v. 13b, but the LXX of v.
14 is closer to the Hebrew of v. 13 than is the LXX of v. 13. For a°71222 is
literally rendered by €v 6uvatots in v. 14, while the comparative state
of Loxupotépous and the pronoun pou in v. 13 reflect 719 of the next verse.
On the other hand, étodéuet of v. 14 reflects the root 17, while
Tatretywodv probably reflects the consonants of MT, though vocalized
differently: 177 (cf. manuscripts bocze2 in 2 Sam 22:48).
Further examples of remote doublets follow:
V. 15b contains only one stich in the Hebrew (7?jn O°77) 721KI N1753
7°), but the Greek contains three:
a tvatt ov Kkatotkets Ev Léow YELAEwV
B 8 退te.vev év Tots toolv avtot
y €v Srarpécets (S8:atpéaeatv) PouBny weydAor dkpiBacuol KapSlas
Only stich y has a counterpart in the Hebrew of v. 15b. The two addit-
ional verses may have been added deliberately, somehow repeating
vv. 16a and 15a, because of a literary principle which may be
recognized by the juxtaposition of these verses.? However, it is doubtful
whether the scribe of codex A was guided by such a principle because
the resultant text makes little sense. It is more probable that the
juxtaposition of the verses is haphazard, even though the wish to
8 See The Codex Alexandrinus in Reduced Photographic Facsimile, | (London 1915).
? Both 15baBy and 16ab contain a group of three stichs in which the first one reflects
oonowan—an)>, and the third one, 2»—n1552. In both cases, the middle line has a different
content. Note further that all three stichs 15baf'y contain a word which denotes a part of
the body: xetdéwv, toalv, kapdlas.
530 CHAPTER THIRTY-SIX
preserve translation doublets in the manuscript tradition was conscious.
The following two translation doublets may be recognized:
(a) l6aa = An? naw? Ta ondwnn
l5ba = tvatt oatKatoukets évpéow XELAE WV
léaa tvatti wporKkd@noat dvd pécov TUV LoodabaLu
l6aa is the more free and hence probably original rendering of the
Hebrew (following de Lagarde’s famous rule),10 and 15ba is secondary:
(1) pot in 16aa does not reflect any element in MT; (2) original
transliterations of rare and difficult words such as o7nown resulting from
the translator’s ignorance were often replaced secondarily by Greek
words.!! The fact that 1éaa is transmitted in the margin of manuscript z
as the fifth Hexaplaric column (0’) seems to militate against the
assumption that this verse reflects the original form of the translation;
however, the precise nature of Hexaplaric quotations needs to be
investigated in full (see n. 5 above), and thus the mere agreement with
the Hexapla does not determine the secondary nature of the quoted
text.
(b) 15ay now ese
1l5ay éEatéotetdkev treCovs avtov
15bB éCETELVEV Ev TOs Trooly auTot
As in the previous example, the rendering ad locum is less literal than
the other one. €v Tots tootv avtov of 15b8 reflects MT 17°7172 more
faithfully than me€ovs avtot. The latter rendering derives the Hebrew
from °*217 rather than from 77;!4 it disregards the bet of 15372 and
therefore is less faithful. Both renderings reflect the vocalization n>w
(thus also B dréote.dev) instead of MT now.
The assumption that the rendering in 15b8 is secondary may be
supported by the fact that €€éte.vev in 15bf is transmitted also separa-
tely as 0’0’ (see, however, above, on @).
More problematic examples follow:
14bB éxetOev Ev oxHTTpW EviaxvovTtos yhoews
The Greek text makes little sense: ‘From there with the sceptre of one
who strengthens of the guidance (?).’ From a grammatical point of view
the text is incorrect because in classical Greek évicytw is used either
10 De Lagarde, Proverbien, 3.
1] ce. Tov, “Transliterations.”* In our case, O‘nDwnn was derived from 75wv represented
in the LXX by xethos.
12 Also elsewhere ~4)7 is rendered by té{os (2 Sam 8:4, 10:6, etc.). For a reverse inter-
change of 237 and *19, see Jer 12:5 o-5:n - ol 1d8es.
THE SONG OF DEBORAH 53:1
transitively as ‘to strengthen’ or intransitively as ‘to prevail,’ and it is
not construed with the genitive (also not in the LXX). The correctness of
é€VLOXvoVTOS in this position is also suspect for an additional reason:
every word in the Greek verse has its counterpart in MT except for
évLoXvoVTos: €kelOev represents O(>)Dwn (reflected as Own), Ev OKT TTPUW
reflects v2nwa, and the rare word tyyhoews reflects 150 (cf. &unyhoews in B
and 6’; &tnyéouar often renders 750 in the LXX).}8
As a result of these doubts with regard to €vioxtovtos, it may be
suggested that this word either forms a second translation of O°Dwn or
represents Ww) of v. 15. The former possibility is not very likely because
the words carry different meanings and consequently évicxvw in the
LXX does not render any verb like ‘to draw.’ The one attractive aspect
of this suggestion is nevertheless the fact that the word is transmitted
in o’ and in several manuscripts of the LXX in its plural form
éviaxvovtes (cf. o°Dwn), but this grammatical form may have been
harmonized contextually.
An alternative explanation of éviaoxvovtos is that the Greek word
reflects “yw) of the beginning of the next verse since that word is not
represented in the LXX in its expected position. This assumption is very
plausible in view of the fact that the root 74w has been rendered in both
Gen 32:29 and Hos 12:3(4) with évioxvw (cf. further below).
Mads Edpaty in v. 14 is not difficult from a linguistic or contextual
point of view (cf. €v Aaots cou at the end of the verse), but these words
do not have any counterpart in MT in the beginning of v. 14. For one
thing, "12 cannot be considered as its counterpart since that word is
represented by pou and the comparative state of laxupotépous at the
end of the previous verse (cf. €€ €ov in 14a, b in A and B for similar
cases of a wrong understanding of *1n). Nor can Aaés be conceived of as a
doublet of 739 since oy and 71 are graphically not similar. Therefore it
may be suggested that \ads reflects oY occurring in one of the adjacent
verses. Indeed, ov in the preceding verse is not represented in A,14 and
its Greek translation may have been added secondarily in the wrong
place.!°
13 @ further possibility is that ty{cew" reflects -1w1 of the next verse (cf. ww - tyyepuiv
passim in the LXX). In that case, 150 is not represented in the LXX,.
4 The word division of MT oY o-17N> is difficult. B reflects 717° oY O°7°7Kx>, while A
reflects an understanding or text like 7177 17°7x(°).
° Thus Schreiner, Septuaginta-Massora, 70. In codex z, \aés is transmitted as 0’6’, and its
inclusion in codex A could thus have derived from the Hexapla. However, the word is
quoted in z as part of the whole phrase (6 Aads Edpatu €TLLWotoaTo avTobs év KOLAGBL).
See further n. 5 above.
532 CHAPTER THIRTY-SIX
evppatvoué vw in v. 11 has often been explained as reflecting a variant
o-nnwn, but that word is graphically not close to MT o7aKXwn.
Alternatively it is not mmpossible that the Greek word derived from a
contextual guess which has little to do with the contents of the Hebrew
word. It seems more probable, however, that evdpatvopévwv reflects
o*pnxn!® which would be a doublet of the difficult o-yxnn of MT, entering
the text at the wrong place.!” This assumption may be supported by the
Hexaplaric reading evdpatvopévwy (o’) which, according to manuscript
z refers to dvakpovopéevuv, the present equivalent in manuscript A of
EPR SN:
The rendering of ninnx by Bactdets in v. 6 may be explained in
various ways. BaotXets may reflect a variant 0°92» (or sim.) which
would create a pair nims - 075n instead of the repeated mnnx in MT. If
Bao.rets does not represent such a variant, it may reflect a second
translation of 71175 of the next verse (cf. dépafwy in A), similar to Suvatol
in B (and to Svvaordv in Hab 3:14 where it renders rm» [K]).18
Particularly difficult is v. 12 where the additional elements form both
regular and remote doublets:
a éEeyelpou éEeyelpou AcBBwoa
Wake up, wake up, Debbora,
B 退yetpov pupiddas peta aod
wake up thousands with a people.
y €é&eyetpou éEeyelpou Addet pet’ WSS
Wake up, wake up, talk with a song.
& évioxvwv éEaviataco Bapak
Strengthen and stand up, Barak,
€ Kal €vLoyvoov AeBBwoa tov Bapak
and Debbora, strengthen Barak.
"ty is the main verbal form in the Hebrew text, but €E€eyelpw even more
so in the LXX because it occurs five times there as against four
occurrences of yin MT. The second verb in v. 12 is €vioytw which has
no direct counterpart in MT. Both verbs are used similarly, each
occurring twice in adjacent lines, once intransitively and once
transitively, while the second lines (8,€) contain the additional
elements. The recognition of this pattern is necessary for a correct
analysis of the difficulties in this verse.
16 evdpalvouat reflects the closely related pnw in Prov 8:30, 31; 31:25.
Thus several commentators. See, e.g., Schreiner, Septuaginta-Massora, 92-93.
18 Thus P.A. Vaccari, Studii critici sopra le antiche versioni latine del Vecchio Testamento, I Il
Cantico di Debora (Roma 1914) 8.
THE SONG OF DEBORAH 533
In the expanded colon of MT,!9-n1y repeated four-times, occurs
intransitively. Possibly this use was problematic to the translator who
also tried to understand the first two occurrences of *1)v as being
transitive.29 This double understanding of *71y is now reflected in a
doublet 7227 ‘Vv parallel to 771129 *y which repeats *1v and contains a
word 7227 graphically very close to 17127. However, the status of peta
daov is less clear. teTd could be a doublet of petd in line y, and Aadés
could reflect a rendering of ov in v. 13 which is absent in the LXX of that
verse, although it is added in v. 14 (see above). However, a solution
based on these lines may be somewhat far-fetched. pupiddas peta daod
may also be reconstructed as ay m22 (for which cf. Ps 3:7),21 where ov
was interpreted as both ay and oy (as in Deut 32:43 iy - weTa Tot Aaov
auTov).
Evioxvwy in line 6 is problematic because it has no counterpart in MT.
The variant €v loyvt of dgnptvw Eth La (in virtute) either derived
from this reading or vice versa. The latter possibility is preferable
because its relation to MT can be determined:* év loxtiit probably forms
a doublet of pet’ wats, deriving 1w from the root mw. A similar exegesis
has been recognized in the OG in v. 14 (see above) and may also be
recognized in v. 29: a°ninWw -loxyvos avtijs (found in an addition of
LXXLUC in v. 30 and virtutis eius of La).2?
Line € may be conceived of as either an exegetical gloss or a doublet.
As an exegetical gloss it matches the first stich of v. 13 in the LXX (but
not in MT). If understood as a doublet, it contains two words (AeBBwpa,
Bapak) which occur in the immediate context, while the third one,
évloxuoov, remains problematical: that word can hardly be explained
as a doublet of €v loyul / €viaytwv (8) which may be a doublet itself.
Alternatively, €vlaxucov may reflect a variant j1n, as many commen-
tators believe, or an exegetical rendering of O1p, since both €vicxtw and
loxus play an important role in the translator’s exegesis in this
chapter:
19 Cf S.E. Loewenstamm, “The Expanded Colon in Biblical and Ugaritic Verse,” JSS 14
(1969) 176-196.
For a similar doublet caused by a wrong understanding of an expanded colon, see the
LXX of Ps 29(28):1, and N.A. van Uchelen, “De LXX-Interpretatie van Ps. 29,” NThT 24
(1970) 173.
Tit is noteworthy that the LXX of Ps 3 also contains other elements which are found in
the pis in v. 12: €mavlotavrat (v. 2), €EnyépEny (v. 6), dvdota (v. 8).
22 Th that case Evioxuwy of A derived from contextual harmonization.
La, together with Exxtes, preserve many original Greek readings; see the literature
quoted in Tov, “Lucian,”* n. 32. Some evidence for Judges was collected by A.V. Billen,
“The Old Latin Version of Judges,” JTS 43 (1942) 140-149.
534
11
(12
is
13
15
25
29
CHAPTER THIRTY-SIX
WWD -
an?) ;
PIR? -
o22 -
"11 (probably) -
alana: -
mnnw -
évlaxvoav
EVLOXUWV)
1} layus avtod
Tous LayupoTté pous
eviaoxvuovtos (14)
Loyupav
Loxvog autrs (see above)
Some ‘regular’ doublets in A are mentioned here:
3
Px)
28
p-117 - oatpatrat duvatot
(cf. B catpdtat and Prov 31:4 Suvdotat).
N11 TNT ON - veavlEwv oLpouaotav avydOn kal olpopdotms.
dvid6y was corrupted from édv 87) (B).
o-11012 - €v payntais dSvuvatots.
MwWRI-TDpPwi - &LexuTTEv—DLdapa 2°
(see Schreiner, Septuaginta-Massora, 93).
The relatively large number of regular and remote doublets analyzed
above shows the complicated history of the transmission of Judges 5 in
codex A, especially in vv. 12-16. This situation may be indicative of
the wide use and frequent copying of this chapter.
CHAPTER THIRTY-SEVEN
THE ‘LUCIANIC’ TEXT OF THE CANONICAL AND APOCRYPHAL
SECTIONS OF ESTHER: A REWRITTEN BIBLICAL BOOK
The so-called Lucianic (L) text of Esther is contained in manuscripts 19
(Brooke-McLean: b’), 93 (e2), 108 (b), 319 (y), and part of 392 (see
Hanhart, Esther, 15-16). In other biblical books the Lucianic text is
joined by manuscripts 82, 127, 129. In Esther this group is traditionally
called “Lucianic’ because in most other books it represents a “Lucianic’
text, even though the ‘Lucianic’ text of Esther and that of the other
books have little in common in either vocabulary or translation
technique. ! The same terminology is used here (the L text). Some
scholars call this text A, as distinct from B which designates the LXX.?
Brooke-McLean? and Hanhart, Esther print the LXX and L separa-
tely, just as Rahlfs, Septuaginta (1935) provided separate texts of A
and B in Judges.
Despite the separation between L and the LXX in these editions, the
unique character of L in Esther was not sufficiently noted, possibly
because Rahlfs, Septuaginta does not include any of its readings. Also
1 Scholars attempted in vain to detect the characteristic features of LXXL"S in Esther as
well. For example, the Lucianic text 1s known for substituting words of the LXX with
synonymous words, and a similar technique has been detected in Esther by Cook, “A Text,”
369-370. However, this criterion does not provide sufficient proof for labeling the L text of
Esther ‘Lucianic,’ since the use of synonymous Greek words can be expected to occur in
any two Greek translations of the same Hebrew text. Furthermore, the tendency of
Atticism, which is characteristic of the Lucianic recension, has been recognized by
Hanhart, Esther, 89 also in the L text, of Esther but the evidence is not strong. For other
characteristics of the L text, see Hanhart, Esther, 87-95.
2 Thus Moore, “Greek Witness” and Cook, “A Text” on the basis of earlier editions. In his
commentaries on Esther and the “Additions,” Moore employs the abbreviation AT (A
Text).
In distinction from the principles used elsewhere in the Cambridge Septuagint, the
edition of the L text of Esther is eclectic, reproducing P.A. de Lagarde, Librorum Veteris
Testament: canonicorum pars prior graece (Gottingen 1883). The L text of Esther has been
printed as a separate text ever since the edition of Esther by Usserius (London 1655).
536 CHAPTER THIRTY-SEVEN
HR does not quote the readings of L in Esther, since it includes only A, B,
S, and the Sixtine edition.4
The L text differs greatly from MT in omissions, additions, and
content. An analysis of its nature is of importance for understanding the
Greek translation(s) of Esther and possibly also for the textual and
recensional history of the Hebrew text. Three explanations suggest
themselves: (1) L reflects a recensionally different text of the book of
which the Greek translation is a reliable, though not a literal,
translation; (2) L is an inner-Greek rewriting of the biblical story. (3) L
is a Greek translation of a Hebrew (or Aramaic) rewriting of the story.
By implication, the assumption of a recensionally different book (1)
bears on our understanding of the literary history of the biblical book of
Esther, while (2) and (3) bear only on the history of its interpretation.
Although the original language of L cannot be determined easily,
possibilities (2) and/or (3) are to be preferred.
I. The dependence of L upon the LXX
L is closely connected with the LXX of Esther and even depends upon it
as shown by idiosyncratic common renderings and errors in L depending
on the LXX. While Hanhart, Esther, 88 demonstrated the close
relationship between the two texts with examples from Additions C
and E, we turn to the canonical sections of Esther.
Examples of renderings common to L and the LXX:
1:20 jo? WWI 2171?
LXX &10 RTMXOD EMS TAOVGIOD
L ATO TTWYUVY EWS TAOVOLWY
9:3 OTT ne ORwID
LXX ETIL@V TODS *lOovdaiovcg
if ETLuwV Tous ’lousalouc
4 Hanhart, Esther, 90, n. 1, mentions a hand-written concordance of L (without Hebrew
equivalents) by P.H. Daking Gooderham (1957).
Thus Moore, “Greek Witness,” Cook, “A Text”, C.B. Paton, Esther (ICC; Edinburgh
1908) 38, and C.C. Torrey, “The Older Books of Esther,” HThR 37 (1944) 1-40. Torrey
described in detail the importance of the two Greek versions of Esther which, in his view,
reflect Greek translations of the original Aramaic text of Esther, from which MT was
translated and adapted.
The possibility that L reflects a recensionally different text of Esther can be supported by
the omission in L of several elements which are problematic in MT (for example, see some
of the minus elements of L described on pp. 540-541). In that case, L reflects a stage of the
development of the book which preceded the expanded text of MT. However, this view
cannot be supported by the other minus elements in L, nor by the known features of that
text. Therefore, the alternative explanations are preferred.
THE LUCIANIC' TEXT OF ESTHER 537
9:3 72D) WR TDA WY)
LXX = ot BaolALKOL ypaUpateic
Ie ol BactALKol ypaupatetc
10:3 wn 7909 Aw “TAT DIT 7D
LXX 0 d€ Mapdoxaioc diedéxeto Tov Bactren AptagepEnv
L 6 5¢ Mapdoyxatoc &leSé xe TO TOV Bactdéa’ AGoUT pov
(for a similar rendering, see 2 Chr 31:12).
Corruption in L shows its dependence on the LXX text:
9:7-10 oA da Mwy ... 1) NNT Nn WW IAD NN? ...
LXX tov te Papoavvectat Ka AEAgov ... TOVG SEKA VDLODG
AUWav
iB Kal Tov Papoav kal Tov adeApov avtot ... Kai TOUG
5é€xa vlovc Ayav
The name of Haman’s son, AeAgov, was corrupted in L to tév dSeApdv avtoi (i.e.,
Farsan’s brother). This reading makes little sense because all the men listed were
brothers. The corruption must have occurred at an early stage because
subsequently the Greek context has been changed: since the list starts by
mentioning “Farsan and his brother,’ it could not any more have the summary line,
‘the ten sons of Haman.’ Therefore, by the addition of kai which is necessarily
secondary since it depends on the corruption of Aehouv to d5eApév, the six names®
were separated from the next phrase —’and the ten sons of Haman.’
8:17 OTN PANT “yA OAT
LXX Kat HOAAOL THV EOVOV MEPLETELOVTO
ie Kal TOAAOL TY I ouSalwy TE pLtEeTELOVTO
The most simple explanation of 0°77°n» would be that the Gentiles ‘became Jews’
out of fear of Mordecai. It was thus understood by the LXX and L: ‘they were
circumcised.’ According to the LXX, this refers to the Gentiles (€6vn); according to
L, to the Jews, but the latter makes little sense. L’s dependence on the LXX shows
in the verb TepteTépovto which derives from the LXX; tov lovSaluv probably
reflects a second rendering of O°-777nn (p17 ~nyn of MT is not represented in L).
4:8 LXX+ ... 810tt ALQV O devtEepedov TH Baorrel EAGANGEY ...
L+ ... OTL Atay 0 b€uTEpe UW AEAGANKE TH Barrel ...
The separation between the translation of the two elements of the phrase 720? 71wn
(cf. MT 10:3 [not in LXX or L]) in L is secondary. In 4:8, the Greek phrase which
has no counterpart in MT refers to Haman, while in the MT of 10:3 it refers to
Mordecai.
6 L mentions only five sons of Haman as well as ‘Farsan’s brother.’
538 CHAPTER THIRTY-SEVEN
The few instances adduced here and the data apud Hanhart, Esther, 88
and B. Jacob, “Das Buch Esther bei den LXX,” ZAW 10 (1890) 261,
demonstrate L’s dependence upon the LXX.’” The exact relationship
between L and the LXX (and MT) is discussed in the next section.
Il. The relationship between L and the LXX
Taking into consideration significant agreements and disagreements
between L and the LXX, we cannot avoid the conclusion that L reflects a
revision of the LXX, as was suggested by most scholars. We focus,
however, on the many differences between L and MT.
Since L is based on the LXX, the many deviations of L from MT must
have resulted either from the translator’s free attitude to his Hebrew
and/or Greek Vorlage or from a different Vorlage. Prior to this analysis
it should be examined whether L had independent access to a Hebrew
text different from MT. That this was the case is evident from the many
syntactic Hebraisms in short additions to MT. Here are some examples
of L, tentatively retroverted into Hebrew:
3:5 + Kal dpyr é€exavbn év abta wal EChtet dvedetv Tov
Mapéoxatov kal tavta Tov Aaov abtot év Yépa urd
= 3nxX ara wy 4D (nx) 72770 AR INA? wpa 1D AIA (nnn) AM
Cf. 2:21 LXX kal €Chtovy dtoKxtetvat = (79n2) 7 mow? wpa and further 1:12
(below).
6:4 (2) + el¢ TapadvdaKhy THs BuxXT|S pou = wD] Nnwn?
6:4 (2) + §L6Ti abvtoc étrolno€ ye Chv axpr Tod viv
=57n/ ane Woe Rik
Note the representation of the hiph4l by to.éw + inf., frequently found elsewhere
in the LXX; see Tov, “Hiphal’™.
6:5 (3) + €veKELTO yap ddBoc Apav év tote odd yxvotc abtav
= 07772 yan In 352 *D
cf. 9:3. ony DTN IND 4D} 7D
Note the different rendering in L and LXX.
? The relationship between L and the LXX is more complex than is implied here, but our
remarks are limited to the canonical sections. It has been recognized (e.g., Cook, “A Text,”
371) that in the sections which have been translated from a Semitic Vorlage (the canonical
sections as well as some of the Additions), the LXX and L reflect two different translations
(see below), whereas the sections which have been composed in Greek (at least Additions
B and E) relate to each other as two recensions of one Greek text.
THE 'LUCLANIC’ TEXT OF ESTHER 539
6:13 (10) + (ag &€ Eyvw Apav Sti ok Av abtdc 6 S0EaCbpevoc
G\N 51. MapSoxatoc) auvetplBn tf Kapdla avTov
opdSpa (kal petéBade TO TvetpLa abtob év ExdAvaet)
= N90 129 ATaw?
6171) +... kal } kapSla altot mpoc Tov KUpLOV = ‘7-28 1271
TZ +... (kal hHywvlacev Ea€np év td dtrayyéAetv OTe 6
dvildikoc) €v ddBadotc¢ auttic kal 6 bedc ESwkev
avTy 8dpaog év TH aVTHY ETLKaXETOObaL adTOV
= pox ANTI ND Ad Mm OPN Prys...?
The assumption that L is based on a Hebrew text may be supported by
renderings which represent MT more faithfully than the LXXx:
1:3 TIN? MPWw TW) ...
LXX .. KAU TOIG APYOVGLV TOV GATPATOV
L ... Kal ol dpyovTEes Tiv ywouwv kaTda TpdowTrov
avuTouU
112 12 TW NAM) TWN F207 APN
LXX KQL EAVTNON O PacUEvS KAL MpPyiGbN
L édutmOn oddSpa Kai opyr) éEexavOy Ev avai
iei4 Jon 7D ONT
LXX OLEYYUS TOD PAaGLAEMS
I Kal ol dpuvtec TO TPdOWTOV Tot BacLrAEws
27. ak 2101 ANN nD"
LXX KQAAOV TO ELGEL
L KaAn Tw Elder odd5pa kal wpata TH OWeEL
S21 21) PRA OIA ANN
LXX eta S€ taDtTE ESdEaEEV
L Kal EYEVETO LETA TOUG Ad'YOUG TOUTOUG
ELEYGNUVED ...
371 ow 2D yn NOD nk own
LXX KQL ETPWTOBGOPEL TAVTOV TOV PIAWV KVTOD
L Kal €8nKke TOV Opdvov avTot UTEpdvw THY
dtrwv adbtot!9
III. The Hebrew text underlying L
L had independent access to a Hebrew (or Aramaic) text which differed
from MT and it probably revised the LXX towards that text. The nature
8 This phrase is known only from biblical contexts. See LSJ, s.v.
9 For further examples, see Moore, “Greek Witness,” 355-358.
10 For further examples, see C.A. Moore, The Greek Text of Esther, unpubl. diss., Johns
Hopkins University 1965, 51; Cook, “A Text,” 375.
540 CHAPTER THIRTY-SEVEN
of this underlying text is investigated here by turning to the so-called
apocryphal Additions to Esther. The two Greek versions of Esther
contain six major additions, traditionally named A-F, besides many
minor additions. The location of Additions A-F at the end of the
canonical sections by Jerome led to misleading conclusions because
scholars usually did not ascribe these Additions to the translator
himself. Although the ‘canonical’ and ‘non-canonical’ components were
mostly studied separately, the combined investigation improves the
understanding of both the LXX and L.
L differs from MT not only with regard to large additions, but also
with regard to large omissions, inversions, and changes. The question of
the original language of the Additions bears on the issues under
investigation. If they were translated from Hebrew or Aramaic, their
fate is closely connected with that of the canonical sections; however,
if some were originally in Greek, they could have been composed by the
translator himself. In this case, the fate of these additions is closely
connected with the Greek version. Scholars believe that the original
language of Add. A, C, D, F was Hebrew or Aramaic, and that of Add. B
and E was Greek.!!
We submit that the translation of the canonical sections in L and the
so-called Additions should be regarded as one organic unit (thus also
Langen, “Esther,” 255):
1, The canonical sections in L contain several references to the
Additions. For example, in 1:1, kal €yéveto peta Tove Aéyouc ToUTOUG (=
T?NT O°7I77 ANN "1, note the Hebraic diction) was added after
Additions A had been prefixed to chapter 1. kal dente tov beod
(4:11[15]) and we étravaato Eo€np tpogevxopévn (5:1) both refer to Add.
C (similar connections with the Additions are found in the LXX to these
verses, and in 2:20; 4:8).
2. When the Additions were attached to the canonical sections,
there resulted a certain redundancy which still shows in the LXX. In L
this redundancy was avoided by omitting some components of the
canonical text. Presumably, the author of the Hebrew (or Aramaic)
Vorlage of L was responsible for these omissions, just as he was
responsible for other omissions and additions. Since both the minor
additions (for examples, see section 5 below) and the large Add. A, C,
D, F were originally composed in Hebrew (or Aramaic), also the
11 coe Langen, “Esther,” 264-266; A. Scholz, Commentar tiber das Buch “Esther” mit seinen
“Zusdtzen” und tiber “Susanna” (Wirzburg 1892) xxi-xxiii; C.A. Moore, “On the Origins of
the LXX Additions to the Book of Esther,” JBL 92 (1973) 382-393; R.A. Martin, “Syntax
Criticism of the LXX Additions to the Book of Esther,” JBL 94 (1975) 65-72.
THE ‘LUCIANIC' TEXT OF ESTHER 541
omissions vis-a-vis MT derived from that Hebrew (or Aramaic) text,
rather than from the Greek translator. Three examples follow:
2:6 om L. The content of this verse (genealogy and background of
Mordecai) is given in Add. A 2(3).
2:21-23 om L. This section tells of Mordecai’s discovery of a plot
against the king which he subsequently foiled. In the canonical book
this section is of major importance; in L it was omitted, probably
because the matter had already been mentioned in Add. A 9(11)-17.
5:1-2 om L (also in the LXX). Add. D elaborates on 5:1~2 of the
canonical text; hence, the parallel verses in MT were omitted.
Two other omissions concern Additions originally written in Greek:
3:12 om L. The content of this verse is covered by Add. B.
8:7-13 om L. The greater part of these verses were omitted because
they are covered by Add. E which contains the decree which allowed
the Jews to take revenge on their enemies.
3. In a few cases, the Additions share vocabulary with the canonical
sections, €.g.:
A 18 L: kat €¢7Tet 6 Auav kakotrotrjoat Tov Mapdoyatov (similar to
the LXX); cf. 3:5 L kal &¢7ret dvedetv TOV MapdSoxatov (different from
MT); cf. also 2:21 LXX and E 3, Land LXX. ©
A 14 L: kat 6podoytoavtec ol Eetvotvyo. dri yOncav (similar to the
LXX); cf. the use of this verb in 7:11 avay&jTw Apav kat un CATw (differ-
ent from MT).
The ‘canonical’ and ‘apocryphal’ sections of L should be considered
as one unit, although the translation of the latter does not depend on
the former. An illustration is 2:6 which is quoted in A 3 in a wording
which is closer to MT than to the LXX and L of 2:6. A 3 mentions
‘Jechoniah king of Judah,’ as does MT of 2:6, whereas the Greek versions
(or their Vorlage) omitted the words between 77)7 O9 and 7737 Wr,
possibly due to homototeleuton (resulting in the strange construction of
the LXxX).
IV. The nature of the Additions in L
If the canonical and non-canonical sections of L (with the possible
exclusion of Add. B and E)!* indeed formed one unit, we should now turn
to a characterization of the L text as a whole.
12 at least Add. E was probably added secondarily. A short version of the original text of
the letter is found after 8:35 in L (before 8:15 of MT), while an expanded version of that
letter (Add. E) is found after 8:12.
542 CHAPTER THIRTY-SEVEN
1. The additions in foto and in many details are secondary when
compared with MT. The contradictions in content between the
Additions and the canonical text of Esther have been amply illustrated
by Paton, Esther (see n. 7) 43; Moore, Additions, 179.
2. In its revision of the LXX, L often follows the Vorlage closely, and
at times represents it more faithfully than the LXX (see section II). At
the same time, the revision often deviated from its Hebrew and Greek
Vorlagen as is shown by a comparison of the three texts in the following
instances:
1:4 WN? NARNN AP?
LXX KQL Thy S6EQv Tic EDPPOD DVI TOD TAOVTOD ADTOD
L Kal THY TLV THS KauynoEws auTot
a2 JD. °D yoT> omnnwa ayAD J2nA WwwI AWK Jenn Ty 7D)
minnw? x27 yrD x2 va 7-1 7707 1? DE
LXX KQL TAVTES OL EV TH ADAT MPOGEKVVODV ADTA. OUTMS
Ya TPOGETAEEV O BAOLAEVS ToLhoat 0 be Mapdoxaiog
OD TPOGEKVVEL LYUTAD
L TavTwy ovV TpoOKUVOUVTWY alTaV KaTad TO TEPdOTAYLA
Tov BaotrAéws MapSoyatos ov TrpogeKUVEL AUTH
513 qwr TPT DTT Nk ARTS Ter ny 992° mw wx AT 25
Joon vvwa
LXX KQL THOTG LOL ODK GPEGKEL OTAV 16@ Mapdoxaiov Tov
“Tovdatov ev tH avan
L TovTO 5é AuTET LE Ldvov S&Tav l6w Tov MapSoxatov Tov
"TovSatov év TH avAy Tod BaotAéwe Kal py Tpookuvet LE
It seems impossible to conciliate the literal and the free elements in L.
Moreover, the LXX reflects renderings of both types throughout the
canon. Accordingly, their juxtaposition in the L text of Esther is not
surprising. Furthermore, one should pay attention to the proportions.
The non-literal elements in L seem to be dominant. Finally, the literal
elements are found more frequently in the former than in the latter part
of the book (see below).
3. The author of the text which underlies L (or simply L, as he will
be called in the following discussion) felt free to rewrite the biblical
story. He added, omitted and rewrote many details. Approximately
half of the biblical book was omitted in L. While in chapters 1-7 L
followed at least the framework of the biblical text, little was left of
chapters 8-10. The editor appears to have been more interested in the
first two thirds than in the last third of the biblical story. It is of
THE 'LUCIANIC’ TEXT OF ESTHER 543
interest to note here that in Midrash Abba Gurion,!3 chapters 8-10 are
not represented, and in b. Meg. 10b-17a (a running commentary on
Esther), they are commented upon very briefly. Possibly the account of
the Jews’ revengeful killing of their enemies was not to the liking of the
authors of L and of these midrashic collections.
4. The introduction (Add. A) and subscription (Add. F) provide the
framework into which the author integrated the rewritten story. One
of the main features of the text behind the LXX and L is the emphasis
on the role of God behind the events.!4 God informs Mordecai in a dream
(A 4-10) what he plans to do. The meaning of the cryptic dream is
clarified in the postscript (Add. F).!° In the biblical story, Esther is not
concerned about dietary laws when she dines with the king, but in Add.
C 27-28, she is extremely concerned about this issue. D 8 mentions God’s
intervention (kal peTéBadev 6 Beds TO TVEdLA TOU BaoLrAéws [then God
changed the spirit of the king]) and C 20 refers to the temple (oBéoat
56Eav olkou cov Kal Ovotaomptdv cov [to quench the glory of your house
and altar]).
A conspicuous feature of the biblical story is the absence of the name
of God.!6 However, in the LXX and L God is mentioned often, both in the
Additions and in the canonical sections.!” Thus, in a free rendering of
the Hebrew (2:20) of Mordecai’s instructions to Esther, the LXX adds:
doBetobat Tov Bedv kal ToLetv Ta TpooTdypLaTa avTod (to fear God and
execute His commands). A similar addition is made in another
instruction of Mordecai to Esther (4:8): émukdAeoat Tov KbpLov (to invoke
the Lord). An addition of kUptos (the Lord) is found in 6:1, but the
textual status of this verse is unclear. Likewise, in L, Mordecai says to
Esther in 4:14 (9) ddd’ 6 Beds Eotat avtots Bonbds (but God will help
them). There are similar additions of 8€6¢ (God) in 4:16 (11) and in 7:2:
Kal 6 860s ESuKev alTH Bdpoaos (and God gave her courage). Noteworthy
is the translation of 7nx Dipan (from another quarter) in 4:14 as 6 Oedc
and the mention of God in connection with the reference to fasting in
4:16. For the text of 7:2, see section II above.
5. L embellished the story as he saw fit. The subjective nature of
these embellishments precludes any consistency. Such expansions are
found in Add. D (the appearance of Esther before the king, parallel to
136. Buber, Sammlung Agadischer Commentare zum Buche Esther (Wilna 1886) 12.
14 See Moore, Additions, 158-159; W.H. Brownlee, “Le livre grec d’Esther et la royauté
divine—corrections orthodoxes au livre d’Esther,” RB 73 (1966) 161-185.
15 See E. Erlich, “Der Traum des Mardochai,” ZRGG 7 (1955) 69-74.
© For the background of this and related issues, see S. Talmon, “’Wisdom’ in the Book of
Esther,” VT 13 (1963) 419-455.
See Brownlee (n. 14).
544 CHAPTER THIRTY-SEVEN
5:1-2 of MT), Add. B (the first letter of the king, after 3:13 of MT) and
Add. E (the second letter of the king, after 8:12).
The canonical sections contain many additions which are not found in
MIT.e.8 5
12 + OTL HKUPWOEV OvaaoTLV TV BovAny avTot
= WSY NX NWI ADT 7D
cf. 1:16 + OTL HKUpWOE TO TPdOTAYLA TOD BaGLAEWC
= 7207 N¥Y NX WDA 7D
dkupdw does not occur in the canonical books of the LXX. It is known from
Aquila’s revision where it frequently equals 197. Hence, it is plausible to
retrovert here the phrase 7xy 757 which occurs often in the Bible.
4:8 Contains a long addition which is partly based on the
LXX. The mention of anD7 yywnd in MT probably
prompted L to dwell on its assumed contents.
5:14 + €TeEL OUYKEXMPNKE O€ 6 BaoLAEUG ddavloat ToUG
"louSalouc kal €S5uxdv oot ol Geol ele ExSl(knoLv avTav
Tye pav OdAE Pptov
Gi2=3 Long addition. The king stresses that nothing was done
for Mordecai. His servants hesitate to answer him
because they envy Mordecai. L draws a parallel
between Mordecai’s and Haman’s fear (for the text,
see p. 538).
6:10 Short addition, see p. 539.
6:11 Long addition, see p. 539.
Te Long addition, see p. 539.
7 Long addition. Esther pities the king, soothes him and
asks him not to be angry. The king makes Esther swear
to tell him who is the evildoer. The addition adds
much dramatic effect to the story.
6. In the rewriting of the biblical story, the author was guided by his
understanding of its major points, so that details were often omitted.
Again, no consistency should be expected. It could be argued, e.g., that
the genealogy of Mordecai and the historical background as depicted in
MT (2:6, not in L) fit the religious tendencies of L, and therefore should
have been retained. However, L may have omitted the verse because
he considered it of little importance for the main thread of the story,}8
18 The verse presents an exegetical problem, especially when it is compared with 1:1 ff.
If Mordechai was deported with the exile of Jechoniah in 597 (2:6), and if wmiwnr is
identified as Xerxes who reigned from 486 until 465 BCE, Mordechai must have been over
100 years old when the events described took place, and his adopted daughter must have
THE 'LUCIANIC’ TEXT OF ESTHER 545
like many other seemingly less relevant or unnecessary details. Further
examples follow:
13
1:10
1:14
1:17-18
22
2:7
2:8-18
yw nw
In L the symposium is not dated.
(nYAw 0DND) ANT XNIINI XMD RIAN NNT pow)? WrxX
(WITWNY 7207 "ID NX Onwnn) ao on
The names of the seven eunuchs are not mentioned. The
sections in parentheses are missing in L.
(nY2w JDM NIOW OW) W'WAN XNOIN ANY NIWID) PX AWM
"FI 015 “Ww
The names of the seven princes are not mentioned.
om. These verses interrupt the sequence of vv. 16-19.
The suggestion that the rebellion of Vashti could cause
other women, especially princesses, to rebel, is a mere
afterthought.
Omitted. V. 22 contains two elements: (1) the king sends
a letter to all provinces; (2) the gist of the letter is that
every man should rule in his own house and be
permitted to talk in his own language. This verse some-
how continues vv. 17-18 which are also lacking in L.
V. 22a is not needed in the context since v. 21 already
mentioned that the king accepted the advice of
Memuchan.
TWAT) (O87 D8 72 PX 7D) TT ND ANON (X77 OTT) NN POX 77
(n> 19 °D77n ANP? ANNI WAN ND) WAN NWI ANN np
The sections in parentheses are missing in L.
L condensed the long and detailed description. A large
part of the section is lacking in L: 8a, 9b, 10-13, 14b, 15—
16, and small segments of vv. 17-18. As a result, the
ceremony of assembling the maidens is missing, as well
as details of grooming the maidens for their meeting
with the king. L knows that Esther was chosen from
among many maidens (v. 17 we 5€ KaTepavOavev Oo
BaoLi\e€uc tdoac tac Tap évouc). In the rewritten text
(in which vv. 10-13 are lacking), v. 14a 782 x77 TQy2
been too old for acting like the biblical Esther. Moreover, 2:6 contains the only allusion to
the history of the Jewish people. As is well-known, the Book of Esther lacks a religious
background, and contains no references to either Palestine, the temple or Jewish history,
except for 2:6. This underlines the difficulties inherent in 2:6. Therefore L may have
omitted this verse on purpose.
546
2:19-20
4:3
4:4-11
CHAPTER THIRTY-SEVEN
aw N71 7722) refers to Esther (cf. the added phrase
wg S€ elonx8n Eo8np tpdc Tov Baovdéa), rather than
to the maidens in general.
Omitted. Exegetes consider these verses as contextually
very difficult, especially v. 19a (which is also lacking
in the LXX). After Esther has been chosen as queen and
the symposium was held (v. 18), there was no need for a
second assembling of the virgins (v. 19). For this reason,
this section may have been omitted.
Omitted. This verse describes the situation in the
Persian empire. It disturbs the connection between vv. 2
and 4, which concern the personal fate of Mordecai.
The section is much shorter in L. Inter alia, vv. 4, 5-7
are lacking. There also is a difference in subject matter;
according to MT, Esther sends Hatakh to Mordecai,
but in L Hatakh is not mentioned. The section which
reports the sending of messengers is condensed in such a
way that the initiative seems to come from Mordecai.
For further omissions see 4:13b; 5:11, and chapters 8-10,
where little of the biblical story is left.
7. The author felt free to make changes and revise whole sections:
1:13-15
3:1-5
3:6-13
In L, v. 13 is followed by v. 15 and then v. 14. The syntax
of MT in wv. 13~15 is difficult. L gives the only correct
interpretation of this text by connecting the verb of
v.13a with v. 15 (v. 13b as well as v. 14 contain
subordinate clauses). The order vv. 15, 14 probably
resulted from the syntactical rewriting of the passage.
Much of vv. 1-5 differs in L from MT, but the message of
both texts is basically the same. There are several
omissions and additions. Note especially the addition
in v. 5 (mentioned above, p. 538) which reflects Hebraic
diction.
Vv. 6-13 occur in L in the sequence: 6 8 9 11 10 7 13. The
most important result of this change vis-a-vis MT is
that the choice of the 13th of Adar succeeds Haman’s
coming to the king. In a way, L’s sequence is more
logical. Haman would not have chosen the day for the
attack on the Jews before permission was granted by the
king. The sequence in MT has caused some exegetes to
THE 'LUCLANIC' TEXT OF ESTHER 547
explain the throwing of lots as referring originally to
the choice of the day on which it was most suitable for
Haman to come to the king.
3 LS m1 ww Wy)
These words occur in L after ‘and Mordecai knew all
that had happened’ (4:11).
7:10 om. The idea in MT was expressed differently else-
where in L: kal Eodpaylabn Ev avtTd 6 Bloc avTod (7:13).
8:1 om. L does not explicate that Ahashverosh gave
Haman’s house to Esther. Instead, he has the king
complain to Esther that Haman wanted to kill
Mordecai, and adds that the king did not know that
Mordecai was Esther’s relative (7:14).
8:2 According to MT, Esther gave Haman’s house to
Mordecai; according to L, the king himself gave him
the house (7:15).
8:3-6 om. Instead, L has Mordecai, not Esther, asking
the king to annul Haman’s edict (7:16).
8. L reflects midrash-type exegesis of the biblical story, adding and
stressing elements in a way which resembles techniques of the
Targumim, the Genesis Apocryphon and several apocryphal and
pseudepigraphal works. Moreover, L not only reflects this type of
exegesis, but occasionally also agrees with actual midrashim on Esther
in the Targumim and in the collections of midrashim (see n. 13). The
clearest example is the above-mentioned feature of placing the story in
a religious setting. For agreements in details, see 4:15 in L and Targum
sheni, 1:5 owt}pta and Yalqut Shimoni ad loc.. roy ayy oN Ww
anwa qwy) XD yw2Dwd) ywIaD> 72m1 nvD7px; Esther’s concern for dietary
laws in Add. C 27-28 is also attested in b. Meg. 13a, Midrash Panim
Aherim II, 63,64, and Targum sheni 2:7. In all these sources God is the
main agent behind the scene and his existence is felt in all sections of
the book.
In 1:16 L equates 72109 with Bovyatog (LXX: Mouxatoc), the equiv-
alent of ‘the Agagite-—Haman (thus 3:1; 9:10; E 10). The equation of
721 and jon is found also in b. Meg. 12b and Midrash Abba Gurion 1.
The second royal letter, Add. E (after 8:12), has a parallel in a
similar addition in Targum sheni.!? The words 7nx o1pnn are taken to
refer to God in L, Targum rishon and Targum shent.
19 For the frequent agreement of (presumably original elements in) La with midrashim
and the Targumim, see G.A. Moore, Esther, 96-127.
548 CHAPTER THIRTY-SEVEN
9. Like L, the LXX is in the nature of a rewritten story, with large-
scale deviations from MT. Like L, the LXX contains large Additions,
and also minor additions and omissions. However, on the whole, the
LXX does not deviate from MT as much as L (the greater part of chapter
8—10 which is lacking in L, is found in the LXX). The precise nature of
the midrash-type translation of the LXX must be studied separately.
To summarize, L is a translation which is based on the LXX but corrects
it towards a Hebrew (or Aramaic)22 text which differs from MT. This
text was a midrash-type rewriting of the biblical story. Clines and Fox
go one step further since according to them, L reflects a different and
pristine text, which helps us to reconstruct the development of the
book.#! If that view is correct, the L text of Esther is of major
importance for the literary analysis of that book. Clines, for example,
believes that the original form of Esther ended at 8:17 (7:17 in the L
text). Also Jobes believes that the L text of Esther 1s based on a Hebrew
original, much shorter than MT, but very similar to that text where the
two overlap.*4
20 The issue of the language underlying L must be studied in greater detail. Torrey’s
arguments (see n. 5) in favor of an Aramaic Vorlage are not convincing, but this possiblility
cannot be discarded. One is struck, e.g., by the sequence of the words in 1:16 Kai
évetelAato 6 Baoivevs tepl ToU Mapdoxalov bepatevetv avtov ... kal Tdoav BUpav
émipavuse Thpety; 6:17 kal E8éKer MapSoxatoc tépac Bewpetv, which is neither Hebrew nor
Greek, but Aramaic.
21 DJ.A. Clines, The Esther Scroll—The Story of the Story (JSOTSup 30; Sheffield, 1984);
M.V. Fox, The Redaction of the Books of Esther (SBL Monograph Series 40; Atlanta, GA, 1991).
2KH. Jobes, The Alpha-Text of Esther—Its Character and Relationship to the Masoretic Text
(SBLDS 153; Atlanta, GA, 1996). On the other hand, K. De Troyer, Het einde van de Alpha-
tekst van Ester (Leuven 1997) believes that L presents an inner-Greek revision not based on
a different Hebrew Vorlage.
CHAPTER THIRTY-EIGHT
THE SEPTUAGINT ADDITIONS (’MISCELLANIES’) IN 1 KINGS 2
(3 REIGNS 2)
In the middle of chapter 2 in 1 Kings, the LXX (3 Reigns 2) has two long
additions, to be named here as Additions 1 and 2. After v. 35, the LXX
has 14 additional verses, traditionally denoted 353-5, and after v. 46
there are 11 additional verses, denoted 46,.). These are not the longest
additions in the LXX of 3 Reigns, since in 12:24 the LXX adds no less
than 24 verses (12:24,.,), and another long addition follows 1 Kgs 16:28
(28,-h).
These additions were noted long ago and their content has been
discussed in detail. Some scholars claim that the LXX reflects a
Hebrew version of 1 Kings which differed much from MT, while others
ascribe these deviations in the LXX to inner-Greek exegesis. Leaving
the analysis of the text-critical value of the additions to the second
part of this study, we first dwell on their content. These Additions,
often named ‘Miscellanies,’ are characterized by the remarkable
phenomenon that the greater part of their contents recurs elsewhere in
the Greek text of Kings, albeit in a slightly different wording. Because
of these discrepancies, one of the suggested solutions can immediately
be discounted, viz., that the Greek translator repeated and rearranged
fragments of his own translation.
The contents of chapters 1-2 of MT are now reviewed in order to
enable an evaluation of the Additions in their context. The sections
which are common to the MT and LXX in 1:1-2:46 (the last days of
David and the accession of Solomon) are: 1:1-53 (the adoption of
Solomon as heir after the final crisis in the struggle for succession)—
subdivided into: the old age of David (1:1-4), the pretensions of
Adonyah (1:5-10), the counter-claims of Solomon (1:1-40), and the
failure of Adonyah’s attempt (1:41-53); 2:1-12 (David’s final charges
to Solomon, the accession of Solomon as sole king ); 2:13—25 (the end of
Adonyah); 2:26-35 (the elimination of Adonyah’s party). After this
550 CHAPTER THIRTY-EIGHT
section the LXX adds verses 35,., (Addition 1), almost all of which
recur elsewhere in 1 Kings:
a = 1 Kgs 5:9
b =o010
C = 99); -6 08D. (Cia7: 1)
d = 5:29
e cf. 7:24, 38 (11, 24)
f = 162 Zo. 9:24
g =(0:25
h = 9:23; 5:30
j = 95153 7, 16 Che 10:224: (XX)
k =
] = 2:8a
mn = 25
O = 29
After this Addition the common text of MT and the LXX continues with
the story of the death of Shimei (2:36-46), followed in the LXX by vv.
46,., (Addition 2):
a cf. 1 Kgs 5:9; 4:20
b = 51 ef, 26k
e cf. 9:18
d cf. 9:18
e = 5:2-3
f = 5:4
g = 5:4-5
h =4:25,0¢,.37) 4%, 6, 4,5
1 = 5:6
k cf. 46b, 5:1, 10:26a
] =/4-1
Chapter 3 covers various aspects of Solomon’s reign: his marriage to the
daughter of Pharaoh (3:1), the worship on the high places (3:2-3), the
dream at Gibeon (3:4-15), his wisdom in the arbitration between the
two women (3:16—28), his administration (4:1-5:8), and his wisdom
(5:9-14).
The content of the two Additions is analyzed next, especially their
inner logic and connection with the surrounding verses. The relevant
texts are adduced below:
(1) The Greek text of the Additions.
(2) The reconstructed Vorlage of the Additions.
THE LXX ADDITIONS IN 1 KINGS 2 son
(3) Elements of MT when differing from (2).
The Greek text is quoted according to Rahlfs’ edition, while the
retroversion of its underlying text is ours, for the greatest part covered
by parallels in MT and further facilitated by the translator’s
adherence to the source text. References to these parallels are provided
and the differences between the reconstructed text and these parallels
are denoted in a third column. When no differences are denoted, the
reconstructed text is identical to the parallel in MT.
reference (+ parallel) retroversion parallel text (if different)
35a (= 5:9)
KQL ESMKEV wm
KUPLOG a ON
(POVNOLV MON
TO LAAOLOV mw?
KQAL COMLAV mam
TOAATV an
Gpodpa TN
KQL TAGTOC a
KAPStAcG ay
OS TN @ULLOG OND
q WR
TAPE (now) by now oy
TTV OGAaGGayV on
35b (= 5:10)
KQL EmAnOvdven am)
N PPOVOIG npon
LAAOLWV mw
opodpa TWN =
VIED THY PPOVYGLV nnonn
TOVTOV nD
KLPYALWV VIOBV O72 "12
KQL DTEP TAVTAC vaeye)
POVILOVG on noon
ALyontov oe
S9¢(= 5176535)
Kat EAaBEv mp1
-- miaey)
a2
THY OVYATEPA
Papaw
KQL ELONYAYEV MVTNV
ELG TNHV MOAL
Aad
EMC
COUVTEAEGQL ADTOV
TOV OLKOV ADTOD
KQL TOV OLKOV
KDPLOD
EV TPWTOLG
KQL TO TELYOG
lepovoaAnu
KUKAOBEV
(= 6:38; cf. 7:1)
EV ENTE
ETEOLV
ETMOUNGEV
KQL ODVETEAEGEV
35d (= 5:29)
KQL TV
tO LOAAOLOV
EBSOUNKOVTG
YLALOOEG
QOUDOVTEG
CPO
KQAL OYSONKOVTA
YUALOEC
AKATOLOV
EV TO OPEL
35e (cf. 7:24, 38 [11, 24])
KQL EMONGEV
LAAMOLOV
TV OGAAGOAV
CHAPTER THIRTY-EIGHT
na nx
my7p
ae
Ty ON
415
W
nD
nd NX
ma nx)
7
MVR 7D
naan nx
o>owin
2720
yaw
ow
(?) nwy
2)
71)
naw)
oyaw
away
"NWI
pate)
ow)
avis
ahis
472
wry
nmn>w
on me
nad
KQL TO
VEOGTNPLYLATO
KQL TOVG AOVLTTHPAC
TOVDG WEYAAOVG
KQL TOVDG OTVAOVDG
KQL THV KONVIV
TAS AVANG
KQL THY OAAAGGaV
TNV XAAKHV
35f(= 11:27, 9:24)
KQL WKOSOUNGEV
THV OKPGV
KQL TC EMGAEEIC
AVTAC
KaL SuEKOWEV
THV TOALW
Aavrd
35f (= 9:24)
odtwc!
OvyYatNp
Papaw
AVEBQLVEV
EK TTS MOAEMS
Aavid
ELG
TOV O1KOV GOTHG
OV
MKOSOLNOEV
QvTH
TOTE
WKOSOLTNGEV
THV OKAY
35g (= 9:25)
KGL LAAMLOV
CLVEMEPEV
THE LXX ADDITIONS IN 1 KINGS 2
o-yP?o7 nN}
NDT NR)
ahaa p i
o-TN7 nN}
nD12 nNI
9Rn7
o- nN)
nonin
ja”)
xin nx
(2?) non (nx)
pao
“VY DR
TA
ane
NvONT AR
ane)
Tun
DX WO
ia
7
eye.
1 For the equivalent, cf. Mich 3:4, Job 11:15. This reconstruction gives the best meaning
to the Vorlage of v. f, even though in the same verse 1& is rendered by TérTe.
554
TPEIG
EV TO EVIAVTO
OAOKAVTWOELC
KQL ELPNVLKOG
ETL
TO BVOLAGTTPLOV
6)
MKOSOLNGEV
TO KVOLO
KQL EODULLG
EVOTLOV
KVPLOD
KQL OVDVETEAEGEV
TOV OLKOV
35h (= 9:23; 5:30)
KQL ODTOL
OL “PYOVTEG
Ol KABEOTHLEVOL
ETL
TO EPYO
TOD LOAADLOV
TPEIG
XUALASES
KQL EFSQKOOLOL
ETIGTATAL
TOD AMOD
TOV MOLOVVTOV
TO EpPyar
CHAPTER THIRTY-EIGHT
(2) conys wow
mw.
nvy
on w)
phy
naw
TW
mie
ale,
uPpmM
- nyatss
= WR
2
TI
ow)
man nx
TAN}
“7
olNIn
Dy (1x)
nDNA
nnbws
nw ow
O7D9N
MINN ww
oan
oval
own
TDN
35i (= 9:15, 17, 18; cf. 10:22a LXX)
KQL WKOSOLNGOEV
tyv Acoovp
Kat thv Mayda
Ka thv Tatcep
Ka. THY Bavbopav
TNV ETAVO
Kat Ta BacrAad
ia)
Tun TN
179 nN}
TH AN
Fn ra nN
yoy pann
nova nN}
35k?
TANV
LLETO
TO OLKOOOLLTI OAL
OVTOV
TOV OLKOV
TOD KVPLOD
KQL TO TELYOG
lepovoaAnu
KV KA®
WLETO TAVTH
OKOSOLLNGEYV
TAC TOAEIG
TAVTAC
35 |
KQL EV TH ETL
Aavid
Cf
EVETELAQTO
TO LOAMOWLOV
NEYOV
(2:8)
‘TSo0v
LETH GOD
Dewer
VLOG
Inpa
DLOG
OTEPLATOG
TOD TEU
ex XeBpwov
35m (= 2:8)
ODTOG
KQATNPAOATO LE
KATAPAV
OdvVNPaV
THE LXX ADDITIONS IN 1 KINGS 2
jt
“YIN
WIA
ml (nx)
ol
main nx
oowine
1250
a aeala
3
oun nx
TONA
73)
a1
nl
abs
mow ne
RN?
(971)
ah aes al
yann
xn
pire)
eee
nan)
O° VWAAy
N11)
55D
2 35k has no parallel in the LXX. Against the traditional verse division, the first part of
the verse should probably be connected with v. 1. With peta Tara a new sentence Starts.
556
Ev 1] NHEpYe
ETOPEVOLNV
Elc MApELPOAGS
35n (= 2:8)
KGL ADTOG
KATEBQLVEV
ELC OTAVTTV LOL
ETL TOV lopdsavryv
KQL OLOCE
AUTH
KATA TOD KVPLOV
NEVYOV
on (] 2:8)
Et
PAVATWONGETAL
EV POLPALE
350 (= 2:9)
KL VOV
bY
KOWWGNS ADTOV
OTL
avnp
PPOVILOS
Ov
KQL YVOoH
TOUNGELG
AVTO
KQL KATAEELG
THV TOALAV ADTOD
EV CULATL
E1G GOV
46a (cf. 5:9; = 4:20)
KQL Tv
0 Paclredc
LAAMLOV
(PPOVULOG
CHAPTER THIRTY-EIGHT
orn
n>
o-inn
NT)
vig
nN AP>
iia
YIVNI
abe,
‘12
Nd
ON
nar / min
37n2
any)
ON
Vipin
pe
WN
oon
Ink
nvm
WR MX
mwyn
7)
nm
naw NX
o72
(7)D Nw
am
qonn
Tm 2w
oon
qv
THE LXX ADDITIONS IN 1 KINGS 2
opodspa mista)
KL GO@OG yan
(= 4:20)
Kal lovda ean
Kat lopanr INAW7)
TOAAOL oan
opodspa INN
OS 1 OLLLOG nS
nN Wk
ETL THSG PUAGOONS on >y
ElG TATOO 47?
EGOLOVTEG D728
KQL TLVOVTEG oonw)
KQL YALIPOVTEG onnw)
46b (= 5:1; cf. 46k)
KL LOAWLOV Ww)
TV TH
C“PXWV wy
EV TAOQIG Ze)!
Taig PaoAELaic moana
KQL TOG rm =
T POG PEPOVTEG own
SOC mae
KQL ESOVAEVOV may") oqay)
TO CYAAMLOV TDW 8
TOUGAG 2D
TAG TNLEPAG a
ths Cans avtod mn
46c (cf. 9:18)
KAL LAOAWLOV mW 2w)
TNPEATO Onn
SLAVOLYELV nny?
TO SVVAGTEVLATO nova ne
tod Aipavov 7297
DOs
3 Suvaotevyata, a hapax in the LXX and in Greek (LSJ), should probably be taken as
‘possessions,’ rendering Ny in 9:18 as m2ya (thus most scholars after Montgomery [below,
n. 5] 128-129). The next verse, v. c, relates to 9:19 and both cities are not men-tioned in
Add. 1, i, where other cities from the list in chapter 9 are listed.
308
46d (cf. 9:18)
KQL MVTOG
MKOSOLNGEV
THV Oepwar
Ev Ti] EPTLO
46e (= 5:2-3)
KQL TODTO
TO APIGTOV
TO LAaAWLOV
TOLAKOVTO
KOPOL
GELLOGAEMC
KQL E€NKOVTG
KOPOL
GAEvpOD
KEKOTOVIGLEVOD
(]-5:3)
SEK
LOOYOL
EKAEKTOL
KQL ELKOOL
Boe
VOULGOEG
KQL EKATOV
KPOPAta
EKTOG
ELAQOV
KaL SopKadav
KQL OPVIBWV EKAEKTAV
VOLLGQOOV
46f (= 5:4)
Sti
TV
LPYOvV
CHAPTER THIRTY-EIGHT
NT
2
TATN NX
ae a2
779791
(7) DOOR
ae)
Nid
it)
a4
or
7nN
mblelana
THE LXX ADDITIONS IN 1 KINGS 2
EV TAVTL
TEPAV
TOD TOTALOD
ano Pa.
EWS
Page
EV TOGLW
TOIG BaclAEedoLv
TEPAV
TOD TOTALOD
46¢ (5:45)
KQL TV
AVTO
eipyvn
EK TAVTOV
TOV LENDOV AVTOD
KUKAOOEV
(5:5)
KOL KATOKEL
lovda
Kal Iopana
TMETOLBOTEC
EKAOTOG
Wine)
THV ALTMEAOV ADTOD
KQL DEO
THV OVKTV AVTOD
EGOLOVTEG
KQL TLVOVTEG
amo Aav
KL EWC
Brpoapee
TOGA
TOS TMEP OG
LOAOLOV
FD)
ahs)
la
mann
TY (1)
min)
yoo
5075
429
i
marae
a)
ow
vaya
yay
y710n
qw7
ate
INAW"I
non
WR
nnn
119)
nnn
WINN
OOD
o-nw)
7
Ww)
yaw N23
pis
“5
nny
NOD?
599
560 CHAPTER THIRTY-EIGHT
46h (= 4:2, 5, 6?, 3?, 4?, 6, 4, 5)#
KQL ODTOL TOR}
OL CPYOVTEG ow
TOD LAAMLOV TD2W? WR 1? WR
ACapiow Woy
VLOG 72
LasaK PITS
TOD LEPEMC jaD7
(= 4:5)
KaL Opviov (2?) WITN way}
VLOG ia
Nadav 13
OaPXYOV Ww 2Y
TOV EYEOTYKOTOV epabelg
(= 4:6?)
Kat Edpan (7) a7781 qWw-nx)
ETL oY
TOV OLKOV @UTOD sg
(= 4:3?)
Kat Lovpa (2) wow)
YPAUUATEDS 19071 o°750
Kai Baca 2 Dowin
VLOG 72
AXveaAau ? TPN
AVALULVYOKOV yom
(4:4?)
Kar ABi ? win
VLOG 72
ToaB INP OTT
AOXLSTPATNYVOG (?) xox Oy
(4:6)
Ka AXE (7) yynN) O7I7N}
4 Fora reconstruction of some of the names, see M. Rehm, “Die Beamtenliste der
Septuaginta in 1 K6n. 2, 46h,” in J. Schreiner (ed.), Wort, Lied, und Gottesspruch, Festschrift
fiir Joseph Ziegler (Wurzburg 1972) 95-101.
VLOG
Edpat
ETL
TAG COGELC
(4:4)
KOL Bavare
VLOG
Imdae
EN
THS ADAAPXIAG
KQL ENL
TOD TALVOELOD
(4:5)
KaL ZAXOVP
LOG
Nadav
O ovULBOvAOG
46i (= 5:6)
KQL TOV
TO LAAOLOV
TECOAKNAKOVTG
AXVALOLOEG
TOKOOEG
“ULTOL
ELG APLATH
Ka bMdEKa
ALALCLOEG
LILTEWV
46k (cf. 46b, 5:1, 10:26a)
KO TV
C“LOXOV
EV TOOL
TOIG BAGLAEDOLV
OO
TOV TOTALOD
KQL EWC
yT¢
THE LXX ADDITIONS IN 1 KINGS 2
12
NID
by
(?) xwnn
win
=
yi
DY
(?) *nn2n
2M)
(?) -n2an
aon)
is
yn
(2) porn
7
maw)
O°vIIN
28
NIN
O°D10
yaDAN?
wy ow)
as
O77
—
2Wn
72
o-D2n7
10
77
7)
vaalx\
561
Or]
T1271)
iD
aie
7207
562
CHAPTER THIRTY-EIGHT
CALKLOOVAOV onw 5
KQL EMS Wi
OPtw@v 7)
Alyortov on
46 | (= 4:1)
LAAMWOV mw
VLOG j2
Aavid 1
EBaotAEvGEV 92 420
ETL DY
- 2D
lopanar ONTW?
Kar lovda ww" —
ev lepovoaAnp oowirrg _
Three types of material are recognized among the additional verses:
1. Most verses almost verbatim repeat translations found elsewhere
in the LXX of the first eleven chapters of 1 Kings. In these instances for
each verse found in the Hebrew Bible, e.g. 1 Kgs 5:9, the corresponding
Greek text occurs twice, once ad loc. (3 Reigns 5:9 LXX) and once in 1 Kgs
2:35a. Sometimes the two Greek renderings are (almost) identical,
while in other instances the Addition differs from the parallel Greek
version.
2. The additional verses run parallel to verses found elsewhere in
MT, but in the corresponding place in the LXX no Greek translation is
found (cf. n. 14).
3. Verses or parts of verses which have no counterpart in the MT or
LXX of 1 Kings (e.g., v. 354).
In his monograph devoted to the Additions, Gooding, Relics repeat-
edly refers to them as ‘a strange phenomenon.’ The very collection of
these verses, most of which occur also elsewhere in the MT and LXX, is
indeed unusual. Furthermore, it is surprising to find two collections of
similar content next to one another, the first after v. 35 and the second
after v. 46.
Different views have been expressed on the nature of the two
Additions. In order to assess the data, the context of both Additions and
the internal sequence of their components are discussed first.
The first Addition follows the story of the death of Adonyah (2:13-
25), and the elimination of his party (2:26-35). After the Addition, the
text continues with the story of Shimei. At this stage Solomon is not yet
the central figure in the story—as in chapter 3 onwards—and hence the
THE LXX ADDITIONS IN 1 KINGS 2 563
Addition in the LXX after v. 35 does not fit its context, as it presents
Solomon as already being the central figure in the story. Thus, from a
contextual point of view, the greater part of the Addition is
inappropriate. Its central themes are Solomon’s wisdom, marriage,
offerings, officers, and building operations, and furthermore it contains
an introduction to the story of Shimei, parallel to MT 2:8-9. The story
of Shimei itself is presented in MT (and the LXX) in 1 Kgs 2:36-46, so
that the introduction (35,_,) immediately precedes the canonical story
itself.
The latter part of Addition 1 (35,|_,) is thus appropriate from a
contextual point of view, but the main part (35,_,) is not. Furthermore,
most of the details are premature in the context, and they also occur
twice in the text.
Also Addition 2 deals with various subjects: like Addition 1, it
begins with Solomon’s wisdom, and continues with various aspects of
Solomon’s dominion, the extent of his kingdom, and the statistics
regarding his provisions, officers, and horses. In a way, Addition 2 is
contextually appropriate, as in MT Solomon is the main figure from
chapter 3 onwards (Addition 2 immediately precedes chapter 3). Yet
the reader is struck by the inappropriate placing of this Addition, since
many of its elements are premature, and furthermore most of them are
repeated in 4:20-5:6.
Thus, from a contextual point of view, the greater part of both
Additions is inappropriate. On the other hand, the last verses of
Addition 1 (35)_,5 = 2:8-9) are contextually appropriate, as they intro-
duce the story of Shimei.
We now turn to the internal logic of the Additions. After a general
description of Solomon’s wisdom (35,-p = 5:9-10), Addition 1 moves to
Solomon’s marriage to Pharaoh’s daughter (c = 3:1, 6:38b). Next come
three details about Solomon’s building activities: the preparations for
building the temple (d = 5:29), the temple utensils (e = elements in
chapter 7), and a few details concerning other building activities in
Jerusalem (fq = 11:27b). The story then returns to Pharaoh’s daughter (fg
= 9:24). This verse may continue the account of Solomon’s marriage
mentioned in c = 3:1 (see below). The text continues with a listing of
Solomon’s offerings (g = 9:25) and of the number of Solomon’s chief
officers (h = 9:23). Addition 1 then returns to Solomon’s building
activities (i = 9:15, 17, 18), to which a remark is added (k) which has
no counterpart in MT.
The text now proceeds with no break to the introduction to the story
of Shimei (1g [without parallel]; 1g = 2:84, m = 2:8,; n = 2:83; 0 = 2:9).
564 CHAPTER THIRTY-EIGHT
When turning to the sequence of ideas in Addition 1, we first refer to the
juxtaposition of its two main parts, a—k (Solomon) and I-o (introduction
to the story of Shimei). This sequence remains problematical, although
unrelated issues are also juxtaposed in MT.
The main issue in analyzing vv. a-k remains whether or not they
follow a certain scheme. We tend to deny any consistent logic in these
verses, although some principles are discerned. The Addition does not
consist of a summary of Solomon’s activities for such a summary would
be more extensive. Nor does the Addition present an anthology of verses
devoted to one single subject. Two themes for anthologies have been
suggested. According to Montgomery° and later Gooding® in much
greater detail, Addition 1 contains an anthology of verses relating to
Solomon’s wisdom and building activities. G. Krautwurst argued that
the building activities were the central issue of the Additions.”
One of the main arguments against these suggestions is the fact that
the text itself does not provide sufficient clues for any one of them. The
fact that the first two verses of Addition 1 refer to Solomon’s wisdom
does not make the whole Addition into an anthology of verses on
Solomon's wisdom. Furthermore, how does Solomon’s marriage, the list
of his officers and his many relate to the mentioned topics?
A further problem in discerning a central topic in Addition 1 is that
its internal logic is not always evident.
1. The beginning of verse f mentions that Solomon built the x19
(dkpa). Therefore the phrase at the end of that verse (‘then he built
the x12n’) contradicts its beginning. This contradiction, not found in the
LXX (9:9) where the last words of MT 9:24 are lacking, is created by the
juxtaposition in verse f of MT 11:27 and 9:24.
2. According to verse c (= 3:1) Solomon brought Pharaoh’s daughter to
David’s city, which was to be her temporary dwelling (in the
meantime Solomon wanted to finish building his own house and the
house of the Lord). To this text, which appears more or less in the same
form in 3:1, v. c adds (cf. 6:38) ‘in seven years he (vid., Solomon) did
(this) and finished (it).’ In a rewritten text, a prediction of the length
of the building activities is possible. It is likewise possible that the
verses are arranged thus so as to show that Solomon finished his
) J.A. Montgomery, “The Supplement at End [sic] of 3 Kingdoms 2 (I Reg. 2),” ZAW 50
(1932) 124-129, esp. 129.
7 Gooding, Relics, chapter 2, 8. See also his earlier articles “The Shimei Duplicate and Its
Satellite Miscellanies in 3 Reigns I,” JSS 13 (1968) 76-92; “Text and Midrash.”
7 G. Krautwurst, Studien zu den Septuagintazusdtzen in 1 (3.) Kénige 2 und ihren Parallel-
texten (diss. Mainz 1977), esp. 75, 82—see my review in BiOr 39 (1982) 629-631.
THE LXX ADDITIONS IN 1 KINGS 2 565
building activities before bringing his wife to the new house. However,
vv. d-e do not refer to those building activities. Moreover, only verse g
(= 9:25), appearing in the text after Pharaoh’s daughter is brought to
her new house, mentions explicitly that Solomon finished building ‘the
house,’ i.e., the temple. The sequence of verses in the Addition thus
does not reflect the intention which Gooding, Relics, 18-29 ascribes to
them.
3. Due to a textual mishap, verse f (11:27) probably states exactly
the opposite of what it intended to say. The verse which is quoted in
the Addition, 11:27, says that ‘Solomon closed the city of David’ (710%
117 1y nx), while the Addition has Solomon ‘break through’ the city of
David (kat 8tékopev THv TOALY Aautd, probably = 717 Ty nx p75). The
text of the Addition probably resulted from a confusion with the next
word in 11:27 (p15). The same text recurs in the LXX of 10:22a (parallel
to MT 9:15), but here the idea is phrased correctly: Tod teptppdéat Tov
dpaypov TG TOAEWG Aautd.
It is hard to discover a guiding principle behind the sequence of the
verses in Addition 1, and no theme is recognizable. In a way, this is not
problematic, since MT also contains unusual sequences of verses, e.g., the
first three verses of chapter 3. The first verse of chapter 3 (= 35c) refers
to Solomon’s marriage to Pharaoh’s daughter. The second verse of that
chapter states that the ‘people’ were (still) ‘sacrificing at the high
places because no house has yet been built for God.’ The third verse
relates that Solomon loved the Lord, but he sacrificed on the high
places. There does not appear to be any connection between verses 1 and
2. Likewise, what is the connection among 9:23, 24, 25, 26, which, too,
are repeated in Addition 1? V. 23 summarizes the number of Solomon’s
officers. V. 24 jumps to another subject: ‘But Pharaoh’s daughter went up
from the city of David to her own house which Solomon had built for
her; then he built the Millo.’ V. 25 deals with yet another topic,
Solomon’s offerings, while v. 26 (as well as vv. 27-28) refers to
Solomon’s fleet. It should be remembered that Montgomery, who
initiated the use of the term Miscellany for the Additions in the LXX of
1 Kings 2, actually used the same term for the Hebrew text of 4:20-5:14,
9:10-10:29.8
We now turn to details which indicate an editorial intention of some
kind:
1. The fact that verse c (against 3:1 quoted here) does not explicitly
mention that Solomon married Pharaoh’s daughter may indicate that
8 J.A. Montgomery, The Books of Kings (ICC; Edinburgh 1951) 126, 180, 204.
566 CHAPTER THIRTY-EIGHT
these words have been omitted intentionally. As noted by Gooding,
Relics, 70-71, this tendency is also visible in the midrash.
2. The connection between 3:1 and 6:38, made in v. c, shows editorial
design. |
3. Verse k, which has no counterpart in MT, and which stresses that
Solomon embarked on the building of the cities (i = 9:15-18) after he
finished building the temple and the walls of Jerusalem, shows
editorial intervention. This addition was needed since vv. h-i mention
the building of cities in the context of building the temple. Although
MT states that the temple had already been finished and dedicated,
an inattentive reader of this Addition might infer wrongly from the
context that the building of cities preceded that of the temple. That
the sequence of the building activities was important to Addition 1 is
also shown by the addition (to the biblical text) in verse c of év mpuitog
(= 71WN712?), showing that the building of Solomon’s house and the
temple preceded that of the building of the walls.?
4. Verses 35)_) run parallel to 2:8-9, but are introduced by an editorial
remark which was needed in the context, since they were removed from
their original context, viz., David’s last words to Solomon. The text of
this editorial addition may be reconstructed as: 4n9w> 281 -n 717 TWA]
TN?.
5. The phrase 1? 1wx in 4:2 refers to Solomon, mentioned in the
previous verse. In the new context of the quotation of this verse in 46h
the subject had to be spelled out: an>w? 7wx.
Addition 1 thus contains an amorphous collection of verses related to
Solomon (a-k) as well as a second introduction to the story of Shimei (I-
0). There is no common theme. The fact that the location of vv. l-o is not
coincidental may suggest that the location of vv. a—k is not coincidental
either, but no explanation seems to be available. Some editorial
intervention in Addition 1 is apparent.
The analysis of Addition 2 is easier, as it is contextually appropri-
ate. Yet, its constituent elements repeat elements occurring elsewhere
and they are mentioned prematurely. It is not difficult to recognize an
organizing principle behind Addition 2, as it runs parallel to a
complete section in MT (4:20-5:6), whose text it reproduces with some
changes. Addition 2 is thus less enigmatic than Addition 1.
The text starts off, as in Addition 1, with a generalized remark
concerning Solomon’s wisdom, adjoined with a statement about the
well-being of the Israelites (a = 4:20). After a detailing of the extent of
7 According to Gooding, Relics, 8, the additional €v tpwraig shows that the building of
the temple preceded that of Solomon’s house.
THE LXX ADDITIONS IN 1 KINGS 2 567
Solomon’s rule and the people paying duties to him (b = 5:1), there are
details concerning Solomon’s building activities (c—d, cf. 9:18). Verses e—
g continue with chapter 5, viz., 5:2-5. These verses deal with the
provisions consumed by Solomon’s household (e = 5:2, 3), the extent of
his dominion (f = 5:4), the peaceful results of his rule (g = 5:4), and
Solomon’s officers (h = 4:2-6). Verse i of the Addition continues with
chapter 5 (verse 6), mentioning the number of Solomon’s horses. It ends
with general statements about the extent of Solomon’s dominion (k) and
his rule in Jerusalem (I = 4:1).
Addition 2 follows some kind of design as it starts and ends with
general statements about Solomon’s wisdom, the extent of his dominion
(a-b, k), and the tranquillity of his reign (k). The middle section
basically follows the MT of the first verses of chapter 5, expanded from
other sources, especially 4:2-5. Since the first verse (a) expands 4:20,
and since this verse immediately precedes chapter 5, Addition 2
contains an expanded version of MT 4:20-5:6 (against the traditional
chapter division).
This Addition, then, is construed around a segment of MT, and not
around a central theme, as claimed by Gooding. The section is not
midrashic, nor does it focus on Solomon’s wisdom.
Additions 1 and 2 are probably somehow connected. Both start off
with statements about Solomon’s wisdom and both deal with similar
subjects concerning Solomon’s reign, dominion, building activities, and
life. There are few duplications of details, and those that occur in a
way supplement one another. Addition 1 focuses more on the personal
life of Solomon than Addition 2. One of the cities which is left out in
the list in Addition 1 (Tadmor) is mentioned in Addition 2 (46d).
Thus, because of their similarity, Additions 1 and 2 may have
constituted once one unit, now separated by the story of Shimei.
When turning to the background of both Additions, we first focus on
the relationship between the text of the Greek Additions and the
wording of their counterparts in the Greek. The issue which should be
examined especially is whether the two Greek versions are related to
each other. For this comparison it does not suffice to point to just any
similar or identical Greek formulation of the Hebrew source text. Many
similarities are, in a way, coincidental if the same translation
technique and system of equivalents are applied to the source texts.
Small differences are also not indicative of any special relation, since
any translator may have varied his translation equivalents
occasionally. We therefore look for unique agreements between the
568 CHAPTER THIRTY-EIGHT
Greek text of the Additions and their counterparts in the canonical text
of the LXX. Several such renderings are recognized:
1. 35d Gpow = 1 Kgs 5:29 LXX (MT 730). In both places 930 is read
as 220, rendered by dpotc, also occurring in some other verses in the
books of Reigns.
2. 46e ddevpou KeKOTIaVLOLEéVvoU (= P71 1?) repeats the LXX of 5:2.
The rare word kekotavtopévou occurs in both places, elsewhere
appearing only in Dan 7:7 (= pp).
3. Note the following special equivalents:
Bon.= 9:23 oman = 6- émLoTdtat
30m: = 2:8 nxzqni = - dé&uvnpav
46e = 5:3 "y7 - VOWLABEG
46e = 5:3 2x(9) - €Adgwv (note the plural)
46e = 5:3 "261 - Kal S0pKkddwv
46g =5:5 moa? = - me ToLO6TEG
4. 46e kal dpv(Owv ékrexTov = 077272), reflecting etymological
exegesis of 1272 from the root (1)72 similar to the rendering of 0°x72
with ékAekTol earlier in the verse, is found in both Addition 2 and in
the LXX ad loc.
These unique similarities prove that the Additions were probably
rendered by the same person who translated the main text of 1 Kings.
This translator was not always consistent, since some differences in
translation equivalents are recognized,! but this is a common feature
also elsewhere in the LXX. The text of the Additions has not been taken
from the Greek text of the parallel verses, since there are too many
differences between the two sets of verses, showing that different
Hebrew Vorlagen are involved.
The Additions were originally written in Hebrew, as is evident from
the translation of the canonical verses, and also of some elements
which have no counterpart in the LXX. In those verses, Hebraistic
renderings can be detected which make the possibility of a Hebrew
origin very likely (cf. TCU, 83-85):
35k ATV peta TO olkodSoptioat abtév - N13 7ANX PA
351 év tH ETL Aaurd Ch -°n 17 TWyA}
46g Kal Av abtd elon éx tavTw Tov pepwv abtod KuKAdOEV - 19 AM
Y20 Pay 757 aY?w.
The Additions are indeed a ‘strange phenomenon,’ especially
Addition 1, and it is questionable whether they comprise literary units.
10 35 KaTéBatvev (17°), 2:8 kaTéBn; 35n El, 2:8 ele; 35n kata ToD Kupfov (‘13), 2:8 év
Kuplw; 350 dpdvinog (05n), 2:9 coddc.
THE LXX ADDITIONS IN 1 KINGS 2 569
Of the scholars who studied the Additions in detail, Hanel,!!
Montgomery, “Supplement,” Gooding, Relics, and Krautwurst (n. 7)
recognized common themes, but the former two denied literary unity.
Hanel, 76, is rather extreme, since he speaks of a “Variantensammlung,’
a conclusion which comes rather close to that of Trebolle (“compilacion
de textos hebreos fragmentarios’).!2 Montgomery, “Supplement,” talks
about ‘supplementary material,’ ‘supplement,’ ‘miscellaneous mate-
rial,’ and ‘compilation.’ He recognized the ‘loose connection’ between
the verses, indicating ‘that we are dealing with fragments from the
text that exists in Kings, and we gain no light on the earlier status of
the material.’ Also Gooding, Relics, 106, accepts this loose connection:
‘the miscellanies are collections of alternative translations, variant
readings, glosses, doublets and the like.’ At the same time, he adds ’...
but what is special about the miscellanies is ... and also that the items
have been carefully edited and worked up into themes (see chapter 2).’
However there is too little positive evidence in favor of such a ‘careful
editing,’ even though elsewhere in 1 Kings the LXX possibly reflects
such midrashic exegesis, as suggested by Gooding in a long series of
articles.13 Consequently, we cannot accept the view of Gooding, Relics,
106, that the miscellaneous material has been ‘worked up into two
themes,’ and afterwards ’... have been inserted into the running
narratives as paragraphs in their own right.” The latter view is
problematical, since the first of these ‘paragraphs’ has no intelligible
connection with its context. Gooding, Relics, 107, further states: “they
are carefully arranged [my italics, E.T.] so that their very arrangement
serves the purpose of making some midrashic point. They are the work
not merely of a textual critic but of a Biblical expositor.’ It seems that
these claims are not supported by the evidence, although some
editorial intervention has been recognized.
The evidence leads to somewhat nihilistic views, based on the
conviction that at least Addition 1 does not form a literary unit. Since
the Additions contain some verses which are lacking in their natural
place in the canonical parts of the LXX,!4 and others are found in
11). Hanel, “Die Zusatze der Septuaginta in I Reg 2, 35a—o und 46a-1,” ZAW 47 (1929)
76-79, esp. 76.
12). Trebolle, “Testamento y muerte de David,” RB 87 (1980) 87-103, esp. 101. Further
references are mentioned there and in id., Salomon y Jeroboan, Historia de la recension y
redaccion de 1 Reyes, 2-12; 14 (Bibliotheca Salamanticensis 3; Salamanca/Jerusalem 1980),
esp. 278, 321.
3 For references, see Gooding’s summarizing article “Text and Midrash.”
14 Add. 1 v. fg = MT 9:24b
Add. 1, v.9 = MT 9:25
570 CHAPTER THIRTY-EIGHT
different parts of the LXX,!9 it is not impossible that these verses were
collected. This assumption leads to the idea of a ‘Variantensammlung’
suggested by Hanel (n. 11). This suggestion refers only to a small section
of the verses of Additions 1 and 2. Furthermore, it is not easy to under-
stand why these collections were placed at their present locations at
the end of the second book of Reigns (roughly = 2 Samuel). It has been
suggested by Montgomery that 2 Reigns did not end after 1 Kgs 2:35, as
most scholars surmise on the basis of the Lucianic evidence, but after 1
Kgs 2:10 (after David’s death). In the words of Montgomery,
“Supplement,” 125, ‘The history of David having been concluded and
the history of Solomon from c. 3 on not having been taken in hand, spare
folios at the end of the volume were used for registering materials of
the history of Solomon that were of interest to scribes.’ It may be
surmised that these verses were collected at the beginning of 3 Reigns (1
Kings), although it is unclear where exactly that book started (2:11 or
2:36). This assumption is particularly attractive in view of the repe-
tition in Addition 1 (35)_,) of the introduction to the Shimei story (1 Kgs
2:8-9). Since the main body of that story (1 Kgs 2:36-46) was contained
in 1 Kings (3 Reigns), it is understandable that someone repeated the
introduction to the story from 1 Kgs 2:8-9, since the original introduction
was found in a different scroll (2 Reigns = 2 Samuel).
In conclusion, the two Additions reflect some editorial intervention
such as the correct location of Addition 1, vv. l-o (= 2:8-9), as an
introduction to the story of Shimei. Addition 2 reflects an organizing
principle, as it is arranged around 1 Kgs 4:20-5:6. At the same time, the
internal difficulties and inconsistencies as well as the duplications
between Addition 1 and Addition 2 complicate the analysis. It
therefore seems that the two Additions reflect collections of variants
and other miscellaneous material in which an editor intervened
slightly. Some of the details in these Additions are more original than
their counterparts in the main text.!6
Add. 2, v. ag = MT 4:20
Add. 2, v.b=MT5:1
Add. 2, v. fg = MT 5:4
Add. 2, v. gg = MT 5:5
Add. 2, v. hg § = MT 4:3, 4
15 Part of 5:1, lacking in the LXX, but found in Add. 2, v. 8b, also occurs in an addition to
the LXX of 10:26. The same applies to 5:6 (= Add. 2, v. i). 3:1 (Add. 1, v. c) is not
represented in the LXX ad loc., but appears in the LXX of 4:31.
16 Thus Rehm with regard to the list of Solomon’s officers (above, n. 4). Note also the
inexplicable words Wk 1nx of MT 9:25, lacking in 35g.
SUPPLEMENTS TO VETUS TESTAMENTUM
Pope, M.H. El in the Ugartic texts. 1955. ISBN 90 04 04000 5
Wisdom in Israel and in the Ancient Near East. Presented to Harold Henry Rowley by
the Editorial Board of Vetus ‘l’cstamentum in celebration of his 65th birthday,
24 March 1955. Edited by M. Noth and D. Winton Thomas. 2nd reprint of the
first (1955) ed. 1969. ISBN 90 04 02326 7
Volume du Congres [International pour l'étude de l’Ancien Testament]. Strasbourg
1956. 1957. ISBN 90 04 02327 5
Bernhardt, K.-H. Das Problem der alt-onentalischen Konigsideologie im Alten Testament.
Unter besonderer Berucksichtigung der Geschichte der Psalmenexegese darge-
stellt und kritisch gewurdigt. 1961. ISBN 90 04 02331 3
Congress Volume, Bonn 1962. 1963. ISBN 90 04 02332 |
Donner, H. Israel unter den Volkern. Die Stellung der klassischen Propheten des 8.
Jahrhunderts v. Chr. zur Aussenpolitik der Konige von Israel und Juda. 1964.
ISBN 90 04 02334 8
Reider, J. An Index to Aquila. Completed and revised by N. Turncr. 1966.
ISBN 90 04 02335 6
Roth, W.M.W. Numencal sayings in the Old Testament. A form-cnitical study. 1965.
ISBN 90 04 02336 4
Orlinsky, H.M. Studzes on the second part of the Book of Isaiah. — The so-called ‘Ser-
vant of the Lord’ and ‘Suffering Servant’ in Second Isaiah. — Snaith, N.H.
Isaiah 40-66. A study of the teaching of the Second Isaiah and its consequences.
Repr. with additions and corrections. 1977. ISBN 90 04 05437 5
Volume du Congres [International pour l'étude de lAncien Testament]. Genéve
1965. 1966. ISBN 90 04 02337 2
Congress Volume, Rome 1968. 1969. ISBN 90 04 02339 9
‘Thompson R.J. Moses and the Law in a century of criticism since Graf. 1970.
ISBN 90 04 02341 0
Redford, D.B. A study of the biblical story of Joseph. 1970. ISBN 90 04 02342 9
Ahlstrom, G.W. Joel and the temple cult of Jerusalem. 1971. ISBN 90 04 02620 7
Congress Volume, Uppsala 1971. 1972. ISBN 90 04 0352) 4
Studies in the region of ancient Israel. 1972. ISBN 90 04 03525 7
Schoors, A. J am God your Saviour. A form-critical study of the main genres in Is.
xl-lv. 1973. ISBN 90 04 03792 2
Allen, L.C. The Greek Chronicles. The relation of the Septuagint I and II Chroni-
cles to the Massorctic text. Part 1. The translator’s craft. 1974.
ISBN 90 04 03913 9
Studves on prophecy. A collection of twelve papers. 1974. ISBN 90 04 03877 9
Allen, L.C. The Greek Chronicles. Part 2. Textual criticism. 1974.
ISBN 90 04 03933 3
Congress Volume, Edinburgh 1974. 1975. ISBN 90 04 04321 7
Congress Volume, Gottingen 1977. 1978. ISBN 90 04 05835 4
Emerton, J.A. (cd.). Studies in the historcal books of the Old Testament. \979.
ISBN 90 04 06017 0
Meredino, R.P. Der Erste und der Letzte. Einc Untersuchung von Jes 40-48. 1981.
ISBN 90 04 06199 |
Emerton, J.A. (ed.). Congress Vienna 1980. 1981. ISBN 90 04 06514 8
Koenig, J. L’herméneutique analogique du jfudaisme antique d’apres les témoins textuels
d’Isaie. 1982. ISBN 90 04 06762 0
Barstad, H.M. The religious polemics of Amos. Studies in the preachings of Amos ii
7B-8, w 1-13, v 1-27, vi 4-7, vin 14. 1984. ISBN 90 04 07017 6
KraSovec, J. Antithetic structure in Bibheal Hebrew poetry. 1984. ISBN 90 04 07244 6
Emerton, J.A. (ed.). Congress Volume, Salamanca 1983. 1985.
ISBN 90 04 07281 0
Lemche, N.P. Early Israel. Anthropological and historical studies on the Israelite
society before the monarchy. 1985. ISBN 90 04 07853 3
Nielsen, K. /ncense in Ancient Israel. 1986. ISBN 90 04 07702 2
. Pardec, D. Ugantic and Hebrew poetic parallelism. A trial cut. 1988.
ISBN 90 04 08368 5
Emerton, J.A. (ed.). Congress Volume, Jerusalem 1986. 1988. ISBN 90 04 08499 }
Emerton, J.A. (ed.). Studies in the Pentateuch. 1990. ISBN 90 04 09195 5
. McKenzie, S.L. The trouble with Kings. The composition of the Book of Kings in
the Deuteronomistic History. 1991. ISBN 90 04 09402 4
Emerton, J.A. (ed.). Congress Volume, Leuven 1989. 1991. ISBN 90 04 09398 2
Haak, R.D. Habakkuk. 1992. ISBN 90 04 09506 3
Beycrlin, W. Im Licht der Traditionen. Psalm LXVII und CXV. Ein Entwicklungs-
zusammenhang. 1992. ISBN 90 04 09635 3
. Meier, S.A. Speaking of Speaking. Marking direct discourse in the Hebrew Bible.
1992. ISBN 90 04 09602 7
Kessler, R. Staat und Gesellschaft 1m vorexilischen Juda. Vom 8. Jahrhundert bis zum
Exil. 1992. ISBN 90 04 09646 9
. Auffret, P. Voyez de vos yeux. Etude structurelle de vingt psaumes, dont le psaume
119. 1993. ISBN 90 04 09707 4
Garcia Martinez, F., A. Hilhorst and C.J. Labuschagne (eds.). The Scriptures and
the Scrolls. Studies in honour of A.S. van der Woude on the occasion of his 65th
birthday. 1992. ISBN 90 04 09746 5
Lemaire, A. and B. Otzen (eds.). Hostory and Traditions of Early Israel. Studies pres-
ented to Eduard Nielsen, May 8th, 1993. 1993. ISBN 90 04 09851 8
- Gordon, R.P. Studies in the Targum to the Twelve Prophets. From Nahum to
Malachi. 1994. ISBN 90 04 09987 5 |
Hugenberger, G.P. Marnage as a Covenant. A Study of Biblical Law and Ethics
Governing Marnage Developed from the Perspective of Malachi. 1994.
ISBN 90 04 09977 8
Garcia Martinez, F., A. Hilhorst, J.T.A.G.M. van Ruiten, A.S. van der Woude.
Studies in Deuteronomy. In Honour of C.J. Labuschagne on the Occasion of His
65th Birthday. 1994. ISBN 90 04 10052 0
Fernandéz Marcos, N. Septuagint and Old Latin in the Book of Kings. 1994.
ISBN 90 04 10043 1
. Smith, M.S. The Usantic Baal Cycle. Volume 1. Introduction with text, translation
and commentary of KTU 1.1]-1.2. 1994. ISBN 90 04 09995 6
Duguid, I.M. Ezekel and the Leaders of Israel. 1994. ISBN 90 04 10074 |
. Marx, A. Les offrandes végétales dans VAncien Testament. Du tnbut d’hommage au
repas cschatologique. 1994. ISBN 90 04 10136 5
Schafer-Lichtenberger, C. Josua und Salomo. Eine Studie zu Autoritat und Legiti-
mitat des Nachfolgers 1m Alten Testament. 1995. ISBN 90 04 10064 4
Lasserre, G. Synopse des lors du Pentateuque. 1994. ISBN 90 04 10202 7
Dogniez, C. Bibhography of the Septuagint — Bibliographie de la Septante (1970-1993).
Avec une préface de Pierre-Maunice Bogaert. 1995. ISBN 90 04 10192 6
Emerton, J.A. (ed.). Congress Volume, Pans 1992. 1995. ISBN 90 04 10259 0
62.
63.
64.
Go:
66.
ore
68.
69.
70,1
70,2
rae
v2.
73.
74,
7!
76.
77.
Smith, P.A. Rhetoric and Redaction in Trito-Isaiah. The Structure, Growth and
Authorship of Isaiah 56-66. 1995. ISBN 90 04 10306 6
O’Connell, R.H. The Rhetoric of the Book of Judges. 1996. ISBN 90 04 10104 7
Harland, P.J. The Value of Human Life. A Study of the Story of the Flood (Genesis
6-9). 1996. ISBN 90 04 10534 4
Roland Page Jr., H. The Myth of Cosme Rebellion. A Study of its Reflexes in Uga-
ritic and Biblical Literature. 1996. ISBN 90 04 10563 8
Emerton, J.A. (ed.). Congress Volume. Cambridge 1995. 1997. ISBN 90 04 106871
Joosten, J. People and Land in the Holiness Code. An Exegetical Study of the
Ideational Framework of the Law in Leviticus 17-26. 1996.
ISBN 90 04 10557 3
Beentyes, P.C. The Book of Ben Sira in Hebrew. A Text Edition of all Extant
Hebrew Manuscripts and a Synopsis of all Parallel Hebrew Ben Sira Texts.
1997. ISBN 90 04 10767 3
Cook, J. The Septuagint of Proverbs — Jewish and/or Hellenistic Proverbs? Conceming
the Hellenistic Colouring of LX-X Proverbs. 1997. ISBN 90 04 10879 3
Broyles, G. and C. Evans (eds.). Writing and Reading the Scroll of Isaiah. Studies of
an Interpretive Tradition, I. 1997. ISBN 90 04 10936 6 (Vol. J;
ISBN 90 04 11027 5 (Set)
Broyles, G. and C. Evans (eds.). Writing and Reading the Scroll of Isaah. Studies of
an Interpretive Tradition, II. 1997. ISBN 90 04 11026 7 (Vol. ID;
ISBN 90 04 11027 5 (Set)
Kooy, A. van der. The Oracle of Tyre. The Septuagint of Isaiah 23 as Version and
Vision. 1998. ISBN 90 04 11152 2
Tov, E. The Greek and Hebrew Bible. Collected Essays on the Septuagint. 1999.
ISBN 90 04 11309 6
Garcia Martinez, F. and Noort, E. (eds.). Perspectives in the Study of the Old Testa-
ment and Early Judasm. A Symposium in honour of Adam 8S. van der Woude on
the occasion of his 70th birthday. 1998. ISBN 90 04 11322 3
Kassis, R.A. The Book of Proverbs and Arabic Proverbial Works. 1999.
ISBN 90 04 11305 3
Rosel, H.N. Von Josua bis Jojachin. Untersuchungen zu den deuteronomistischen
Geschichtsbuchern des Alten Testaments. 1999. ISBN 90 04 11355 5
Renz, Th. The Rhetoncal Function of the Book of Ezekiel. 1999.
ISBN 90 04 11362 2
Harland, P.J. and Hayward, C.T.R. New Heaven and New Earth Prophecy and the
Millenium. Essays in Honour of Anthony Gelston. 1999. ISBN 90 04 10841 6
VETUS TESTAMENTUM
SUPPLEMENTS
Edited by
H.M.BARSTAD~- PHYLLIS A.BIRD
R.P.GORDON - A.HURVITZ
A.VAN DER KOOIJ- A.LEMAIRE
R.SMEND -J.TREBOLLE BARRERA
J.C. VANDERKAM
H.G.M.WILLIAMSON
[Se
aaa
The Supplements to Vetus Testamentum series
covers the whole range of Old Testament
study, including Septuaginta studies, Uga-
ritic research relevant to the study of the Old
Testament, Hebrew studies, studies in ancient
Israelite history and society, and studies in
the history of the discipline. There are both
monographs and collective volumes, the lat-
ter including the Proceedings of the Trien-
nial International Congresses of the Interna-
tional Organization for the Study of the Old
Testament.
45 Beyerlin, W. Im Licht der Traditionen. Psalm
LXVII und cxv. Ein Entwicklungszusam-
menhang. 1992. ISBN 90 04096353
46 Meier, S.A. Speaking of Speaking. Marking
Direct Discourse in the Hebrew Bible. 1992.
ISBN 9004096027
47 Kessler, R. Staat und Gesellschaft im vorexili-
schen Juda. Vom 8. Jahrhundert bis zum Exil.
1992. ISBN 9004096469
48 Auffret, P. Voyez de vos yeux. Etude structu-
relle de vingt psaumes, dont le psaume 119.
1993. ISBN 90 0409707 4
49 Garcia Martinez, F, A. Hilhorst and C. J. La-
buschagne (eds.). The Scriptures and the Scrolls.
Studies in honour of A.S. van der Woude on
the occasion of his 65th birthday. 1992.
ISBN 900409746 §
§0 Lemaire, A., and B.Otzen (eds.). History and
Traditions of Early Israel. Studies presented to
Eduard Nielsen, May 8th, 1993. 1993.
ISBN 9004 098518
51 Gordon, R.P. Studies in the Targum to the Twelve
Prophets. From Nahum to Malachi. 1994.
ISBN 9004099875
§2 Hugenberger, G.P. Marriage as a Covenant. A
Study of Biblical Law and Ethics Governing
Marriage, Developed from the perspective of
Malachi. 1994. ISBN 9004099778
§3 Garcia Martinez, F., A. Hilhorst, J.T.A.G.M.
van Ruiten, and A.S. van der Woude (eds.).
Studies in Deuteronomy. In Honour of
C.J. Labuschagne on the Occasion of His 65th
Birthday. 1994. ISBN 9004100520
$4 Fernandez Marcos, N. Scribes and Translators.
Septuagint and Old Latin in the Books of
Kings. 1994. ISBN 9004100431
§5 Smith, M.S. The Ugaritic Baal Cycle. Volume 1.
Introduction with Text, Translation and
Commentary of KTU I.1-1.2. 1994.
ISBN 9004099956
$6 Duguid, I. M. Ezekiel and the Leaders of Israel.
1994. ISBN 9004100741
57 Marx, A. Les offrandes végétales dans l’Ancien
Testament. Du tribut d’hommage au repas
eschatologique. 1994. ISBN 900410136 §
58 Schafer-Lichtenberger, C. Josua und Salomo.
Eine Studie zu Autoritat und Legitimitat des
Nachfolgers im Alten Testament. 199s.
ISBN 9004100644
59 Lasserre, G. Synopse des lois du Pentateuque.
1994. ISBN 9004102027
60 Dogniez, C. Bibliography of the Septuagint —
Bibliographie de la Septante 1970-1993. Avec une
préface de Pierre-Maurice Bogaert. 1995.
ISBN 9004101926
61 Emerton, J.A. (ed.). Congress Volume, Paris
1992.1995. ISBN 9004102590
Continued on backflap
ISSN 0083-5889
ISBN 90-04-11309-6
|
4"113091
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11